biomedical engineering Recently Published Documents

Total documents.

  • Latest Documents
  • Most Cited Documents
  • Contributed Authors
  • Related Sources
  • Related Keywords

Ophthalmological instruments of Al-Halabi fill in a gap in the biomedical engineering history

Virus-like particles: revolutionary platforms for developing vaccines against emerging infectious diseases.

Virus-like particles (VLPs) are nanostructures that possess diverse applications in therapeutics, immunization, and diagnostics. With the recent advancements in biomedical engineering technologies, commercially available VLP-based vaccines are being extensively used to combat infectious diseases, whereas many more are in different stages of development in clinical studies. Because of their desired characteristics in terms of efficacy, safety, and diversity, VLP-based approaches might become more recurrent in the years to come. However, some production and fabrication challenges must be addressed before VLP-based approaches can be widely used in therapeutics. This review offers insight into the recent VLP-based vaccines development, with an emphasis on their characteristics, expression systems, and potential applicability as ideal candidates to combat emerging virulent pathogens. Finally, the potential of VLP-based vaccine as viable and efficient immunizing agents to induce immunity against virulent infectious agents, including, SARS-CoV-2 and protein nanoparticle-based vaccines has been elaborated. Thus, VLP vaccines may serve as an effective alternative to conventional vaccine strategies in combating emerging infectious diseases.

Biomedical Engineering Technologies

Biocybernetics and biomedical engineering – current trends and challenges, deep learning models principles applied to biomedical engineering, deep learning models evolution applied to biomedical engineering, ieee transactions on biomedical engineering (t-bme), biomedical engineering and occupational therapy approach in technologies for enhancement human labor and defense abilities, artificial intelligence models applied to biomedical engineering, introduction to cognitive science, cognitive computing, and human cognitive relation to help in the solution of artificial intelligence biomedical engineering problems, export citation format, share document.

BioMedical Engineering OnLine

New Content Item

Celebrating 20 years of Biomedical Engineering Online

Read our special anniversary collection  here . 

  • Most accessed
  • Collections

Simulating impaired left ventricular–arterial coupling in aging and disease: a systematic review

Authors: Corina Cheng Ai Ding, Socrates Dokos, Azam Ahmad Bakir, Nurul Jannah Zamberi, Yih Miin Liew, Bee Ting Chan, Nor Ashikin Md Sari, Alberto Avolio and Einly Lim

Non-invasive parameters of autonomic function using beat-to-beat cardiovascular variations and arterial stiffness in hypertensive individuals: a systematic review

Authors: Jia Hui Ooi, Renly Lim, Hansun Seng, Maw Pin Tan, Choon Hian Goh, Nigel H. Lovell, Ahmadreza Argha, Hooi Chin Beh, Nor Ashikin Md Sari and Einly Lim

Kinematic difference and asymmetries during level walking in adolescent patients with different types of mild scoliosis

Authors: Run Ji, Xiaona Liu, Yang Liu, Bin Yan, Jiemeng Yang, Wayne Yuk-wai Lee, Ling Wang, Chunjing Tao, Shengzheng Kuai and Yubo Fan

The effectiveness of simple heuristic features in sensor orientation and placement problems in human activity recognition using a single smartphone accelerometer

Authors: Arnab Barua, Xianta Jiang and Daniel Fuller

StairNet: visual recognition of stairs for human–robot locomotion

Authors: Andrew Garrett Kurbis, Dmytro Kuzmenko, Bogdan Ivanyuk-Skulskiy, Alex Mihailidis and Brokoslaw Laschowski

Most recent articles RSS

View all articles

How smartphones are changing the face of mobile and participatory healthcare: an overview, with example from eCAALYX

Authors: Maged N Kamel Boulos, Steve Wheeler, Carlos Tavares and Ray Jones

Automated system for lung nodules classification based on wavelet feature descriptor and support vector machine

Authors: Hiram Madero Orozco, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sánchez, Humberto de Jesús Ochoa Domínguez and Manuel de Jesús Nandayapa Alfaro

Biological effects of exposure to magnetic resonance imaging: an overview

Authors: Domenico Formica and Sergio Silvestri

Organ-on-a-chip: recent breakthroughs and future prospects

Authors: Qirui Wu, Jinfeng Liu, Xiaohong Wang, Lingyan Feng, Jinbo Wu, Xiaoli Zhu, Weijia Wen and Xiuqing Gong

Oscillometric measurement of systolic and diastolic blood pressures validated in a physiologic mathematical model

Authors: Charles F Babbs

Most accessed articles RSS

Article Collection International Conference on Aging, Innovation and Rehabilitation Edited by Milos Popovic, Babak Taati

Thematic series Deep learning in biomedical engineering Edited by Robert Koprowski

Thematic series Advanced signal processing and modeling for neuroengineering Edited by Fei Chen, Shi-xiong Chen and Dong-mei Hao

Thematic series Trends in e-health and m-health: technology developments for improved healthcare Edited by Alessia Paglialonga

Thematic series Artificial intelligence in biomedical imaging Edited by Robert Koprowski

Thematic series Advances in medical robotics and automation for surgery and rehabilitation Edited by Dhanjoo Ghista and Kelvin Wong

Thematic series Biomedical engineering and the heart: coronary blood flow, myocardial perfusion, myocardial ischemia and infarcts detection, customized stents and bypass surgery Edited by Dhanjoo Ghista and Kelvin Wong Thematic series Advances in neuroimaging: insights into neurological and psychiatric disease Cross-journal series Edited by Kelvin Wong

Announcing the launch of In Review

BioMedical Engineering Online , in partnership with Research Square, is now offering In Review . Authors choosing this free optional service will be able to:

  • Share their work with fellow researchers to read, comment on, and cite even before publication
  • Showcase their work to funders and others with a citable DOI while it is still under review
  • Track their manuscript - including seeing when reviewers are invited, and when reports are received

Editor-in-Chief

Ervin Sejdic, University of Pittsburgh, USA

Aims and scope

BioMedical Engineering OnLine is an open access, peer-reviewed journal that is dedicated to publishing research in all areas of biomedical engineering.

BioMedical Engineering OnLine is aimed at readers and authors throughout the world with an interest in using tools of the physical and data sciences, and techniques in engineering, to understand and solve problems in the biological and medical sciences. Read more .

Do you have an idea for a thematic series? Let us know!

Thematic series.

BioMedical-Engineering OnLine has published the following thematic series:

International Conference on Aging, Innovation and Rehabilitation

Advanced signal processing and modeling for neuroengineering

Trends in e-health and m-health: technology developments for improved healthcare

Artificial intelligence in biomedical imaging

Advances in medical robotics and automation for surgery and rehabilitation

Biomedical engineering and the heart: coronary blood flow, myocardial perfusion, myocardial ischemia and infarcts detection, customized stents and bypass surgery

Advances in neuroimaging: insights into neurological and psychiatric disease

Our data and software availability policy

Our journal’s policy is for all data, previously unreported software or custom code described in the manuscript to be made freely available for any scientist wishing to use them for non-commercial purposes, without breaching participant confidentiality. All authors must include an ‘Availability of Data and Materials’ section in their manuscript, detailing where the data, software or code supporting their findings can be found. Authors who do not wish to share their data and materials must state that it will not be shared, and give the reason. However, these authors are still expected to make their data and materials available to reviewers on a confidential basis. Please see BioMed Central’s Editorial policies page for more information.

  • Editorial Board
  • Manuscript editing services
  • Instructions for Editors
  • Sign up for article alerts and news from this journal

Annual Journal Metrics

2022 Citation Impact 3.9 - 2-year Impact Factor 3.8 - 5-year Impact Factor 1.310 - SNIP (Source Normalized Impact per Paper) 0.663 - SJR (SCImago Journal Rank)

2022 Speed 8 days submission to first editorial decision for all manuscripts (Median) 166 days submission to accept (Median)

2022 Usage  1,106,250 downloads 682 Altmetric mentions 

  • More about our metrics

ISSN: 1475-925X

  • Open access
  • Published: 25 September 2021

Biomedical engineering and ethics: reflections on medical devices and PPE during the first wave of COVID-19

  • Alessia Maccaro 1 , 2   na1 ,
  • Davide Piaggio   ORCID: orcid.org/0000-0001-5408-9360 1   na1 ,
  • Concetta Anna Dodaro 3 &
  • Leandro Pecchia 1 , 4 , 5  

BMC Medical Ethics volume  22 , Article number:  130 ( 2021 ) Cite this article

14k Accesses

6 Citations

2 Altmetric

Metrics details

In March 2019, the World Health Organization (WHO) declared that humanity was entering a global pandemic phase. This unforeseen situation caught everyone unprepared and had a major impact on several professional categories that found themselves facing important ethical dilemmas. The article revolves around the category of biomedical and clinical engineers, which were among those most involved in dealing with and finding solutions to the pandemic. In hindsight, the major issues brought to the attention of biomedical engineers have raised important ethical implications, such as the allocation of resources, the responsibilities of science and the inadequacy and non-universality of the norms and regulations on biomedical devices and personal protective equipment. These issues, analyzed one year after the first wave of the pandemic, come together in the appeal for responsibility for thought, action and, sometimes, even silence. This highlights the importance of interdisciplinarity and the definitive collapse of the Cartesian fragmentation of knowledge, calling for the creation of more fora, where this kind of discussions can be promoted.

Peer Review reports

Since early 2020, the current SARS-CoV-2 pandemic has challenged all the fields of knowledge, increasing the need for their interconnection. Medicine, science, politics, and more specialized sectors such as biomedical engineering (BME), faced crucial ethical issues, which can no longer be underestimated.

Biomedical engineers design medical devices, raising many ethical dilemmas in ordinary times, which become compelling during such a crisis. The authors of this manuscript had the privilege of different points of view thanks to the President of the European Society of BME (i.e., EAMBES) and Secretary General of the global society of BME and medical physics (IUPESM).

Three pivotal themes emerged in the global BME community:

The dilemma of identifying criteria for the allocation of medical devices

Responsibilities of science and technology

Inadequacy of regulations and norms, which lack universality

This manuscript does not follow the traditional structure of scientific papers (e.g., methods, results etc.), rather it is a critical analysis, revolving around the 3 above-mentioned pillars. The first pillar focuses on the surfacing of ethical dilemmas in times of pandemic (e.g., scarce resource allocation), delving into examples from Italy, where the new decision-making criteria often clashed against the existing constitutional and moral principles during the first wave.

The second pillar retrospectively reports how the medical device sector was affected by the current pandemic, touching on hazardous amatorial attempts of the general public to face the scarcity of resources and the urgent needs, and on the crisis-related challenges that surfaced for manufacturers, exacerbated by the lack of dialogue with decision-makers. In the same pillar, the theoretical debate among science and politics is addressed, referring the case of the “intended use” of a medical device, from two different philosophical perspectives, i.e., substantialism and utilitarianism. The former underlines the fundamentality of substances as ontological categories [ 1 ], suggesting that the intended use of a medical device should always be respected. The latter stresses the importance of maximizing the overall good, authorizing different uses, depending on the relativistic utility or, in exceptional times of crisis, on the emergency.

This invites the reader to a subsequent reflection on the inadequacy of the existing regulations on medical devices, in the third pillar. In respect to this, the manuscript proposes to offer a hermeneutic perspective close to the situational ethics that authorizes negotiations and mediations between the generality of principles and norms, and the specificity of the context.

Overall, the ethical considerations made in this manuscript, should be considered a valuable lesson for the future of crisis management. If and only if ethics and bioethics will be considered as effective support for science and scientists (doctors, biomedical engineers, etc.), the Cartesian separation of knowledge [ 2 ] could be overcome, establishing an interdisciplinary dialogue that involves peoples and emphasizes the public relevance of such issues.

In this way, such dilemmas could be anticipated by establishing a framework that could provide guidance and appropriate methodology to address arising urgent issues, without having to resort to specialists for questions that concern everyone and need a multidisciplinary approach.

Allocation of medical devices: clinical and ethical principles

Due to limited resources, decision-makers have had to compromise among all the potential useful interventions. Also in the past 10 years, the most public National health systems were massively privatized, resulting in a significant reduction of prevention services and a great reduction of Intensive Care Unit (ICU) beds. The rapid spread of the SARS-CoV-2 resulted in an unprecedented need of sub-intensive and ICU beds, which overcame the capacity of the most advanced national healthcare services. Within a few weeks, the available resources (i.e., medical devices, doctors, nurses) proved to be insufficient to cover the care needs of the multitude of COVID-19 patients, beyond the ordinary needs of other patients. Consequently, doctors and healthcare structures ended up pondering and making difficult ethical choices in a short time and identifying priority principles that could guide them.

National [ 3 , 4 , 5 , 6 ] and international [ 7 , 8 , 9 ] ethics committees, scientific societies [ 10 , 11 ], and experts [ 12 , 13 , 14 , 15 ] soon expressed their opinion on the matter [ 16 ], identifying requirements that respect human dignity and fundamental ethical principles, enshrined in the Charters of both National and International Rights [ 17 ].

The case of Italy, the first country significantly affected outside China, is emblematic. According to the Italian Constitution, health is a "fundamental right of the individual" and a "collective interest" (article 32). In addition, article 2 recognizes the personalist principle and the duty of solidarity, and article 3 establishes the principle of equality. Accordingly, the law founding the Italian National Health Services (NHS) (n. 883/1978) prescribes that care must be ensured according to the principles of universality, equality and fairness.

Leveraging on these fundamental principles, the Italian National Bioethics Committee [ 18 ] remarked the need to allocate medical devices and other resources based only on a clinical criterion, and without considering criteria such as age, gender, social attributes, ethnicity, disability and costs, in compliance with the principles of "justice, equity, solidarity". The envisaged method was that of "triage in a pandemic emergency", based on what the World Health Organization defines as "preparedness" (WHO) as a premise, and on two key concepts, i.e., "clinical appropriateness" and "actuality", identified by the healthcare professional on the basis of clinical criteria [ 19 ].

This point of view, which values the person and opposes attempts of objectification in a series of “pre-established criteria” (except for the point of view of the clinician), is well expressed by J. Habermas. In an interview with the newspaper Le Monde, Habermas underlined the inadequacy, in moral terms, of an “objective quantification” of patients and insisted on the essential issue of the “recognition” of the individual: “when addressing a second person (you, you), the other person's self-determination must either be respected or denied, that is, either accepted or ignored” [ 20 ].

On the other hand, the Italian society of anesthesiology (SIAARTI) introduced the identification of an age limit for accessing intensive care in case of necessity [ 21 ]. SIAARTI’s document explained that COVID-19 created a scenario, in which criteria for accessing ICU may be needed beyond the clinical appropriateness and proportionality of care, but also in distributive justice and the appropriate allocation of limited healthcare resources, which means privileging those with the "greatest life expectancy" [ 22 ]. Even if the choice of whom to admit to treatment is a terrible reality, it must be discussed by ethicists and bioethicists to further identify solutions that support medical doctors in taking these decisions. “This is not to deny their clinical authority and responsibility, but rather to urge a commitment to give such question public relevance" so that in the future the "public health perspective" [ 23 ] to follow is clear from the beginning.

Responsible thinking, responsible actions, responsible silence

COVID-19 created a global lack of essential medical devices (e.g., pulmonary ventilators) and personal protective equipment (PPE, such as masks, respirators) [ 24 ]. This led to an unprecedented amount of do-it-yourself (DIY) solutions, which were fomented on media worldwide. Consequently, ordinary individuals started producing PPE at home with 3D printers and everyday materials, and manufacturers converted their production facilities to develop medical devices and PPE.

Unfortunately, although very admirable, this approach is not feasible in critical sectors such as medical devices or PPE, which require postgraduate education, years of experience and deep knowledge of relevant international standards and norms, in order to ensure appropriate levels of safety, efficacy and resilience. Thus, only 7 manufacturers in the world are producing pulmonary ventilators. In fact, during a pandemic, we do not only need to ensure the usual standard of quality, but we should also consider making those devices more resilient, because, if hospitals fail, we will need to safely operationalize these devices in field hospitals, tent-like structures, and any other relevant setting. Therefore, experience, safe-by-design approaches, and the knowledge of additional standards (e.g., military standards) become relevant too.

Minimal scientific evidence exists on how harmful this DIY wave has been, but few facts can be clearly reported referencing major newspapers. In mid-March 2020, manufacturers were called upon to help to tackle medical devices and PPE crisis. Footnote 1 Footnote 2 Many responded, certainly moved by the noblest principle and willingness to help. Unfortunately, learning to manufacture complicated and highly regulated pulmonary ventilators cannot be done in a few weeks. By mid-April 2020, new productions of ventilators were stopped in Spain, Footnote 3 UK suspended orders of BlueSky ventilators Footnote 4 and France followed. Footnote 5 Once again, a virtuous example came from Italy, where the only Italian manufacturer of ventilators (i.e., Siare Engineering International Group l.t.d.) was supported by the Italian Government, which offered 25 highly specialized army engineers, by the former FIAT (now FCA), who supported producing electro-mechanic components, in addition to Ferrari providing electronic components. With this collaborative effort, Siare increased the production of top-of-the-range ventilators from 160 to 500 units per month, respecting the highest quality standards. In conclusion, the rise of useless and potentially harmful DIY approaches to PPE and medical devices could have been easily avoided at the start of the pandemic by decision-makers initially consulting with domain experts, such as biomedical and clinical engineers.

In this regard, the belief that disciplinary competence is to be sectorized and not interconnected, very often, leads to a further separation of knowledge that is detrimental to people. In fact, it could be good practice that politicians had a solid scientific background in order to legislate about scientific matters, above all if they involve public health. On the other hand, scientists should "grovel in the dirt of the city of Romulus" [ 25 ] keeping their related studies as tangible and accessible as possible, and acquire a more solid political culture and a growing awareness of their social role. After all, the relationship between science, policy-making and politics has been controversial since the dawns of civilization: people like Aristarchus of Samos (i.e., one of the fathers of an early Heliocentrism), censored by sectaries such as Cleanthes (i.e., the prince of stoics at that age), or like Socrates, accused, “censored”, and sentenced to death for being “unorthodox” by one of the most open and democratic societies of the times, the Polis of Athens [ 26 ], or like Galileus, who was condemned for radically opposing to the Sacred Scriptures- and Church-approved Geocentric model, are just a few stark examples. Although much progress has been made since then, there is still an ongoing debate among a Weberian distinction between science and politics and a Habermasian and Marcusean dichotomy between the technocratic and decisionist models of scientific advice to politics. Nonetheless, it is evident that we are transitioning towards ever more present democratization of science, and not without associated risks. In fact, how can this be achieved without compromising the epistemic quality of knowledge [ 27 ]?

In particular, the COVID-19-related debate, involving politicians, scientists (especially biomedical engineers) and ethics experts, is based on two distinct currents of thought, referring to two different philosophical matrices, i.e., utilitarianism and substantialism . In fact, some people believe that any kind of emergency-ready response that can make up for the shortage of PPE and medical devices is “better than nothing”, even at the expense of the safety and efficiency normally guaranteed by the standards. This way of thinking is in line with Utilitarianism’s conception of maximizing happiness and overall gains for all the affected individuals. However, the “better than nothing approach” is dubious and is a well-known logical fallacy, that of the relative privation [ 28 ]. This kind of fallacious way of reasoning also justifies the misuse of something that does meet the standards but was intended for completely different purposes. In fact, the intended purpose is key, at least in the world of medical devices, and it is what safeguards the manufacturers in case their products fail if they are used “off label”. In this case, the liability falls with the individual who misused the product in the first place [ 29 ]. This concept is well portrayed by what happened in Harrow, Footnote 6 , Footnote 7 where some nurses, after denouncing their precarious working conditions and the lack of PPE in the fight against COVID-19, had started using bin bags as a replacement of the unavailable PPE. In this case, the beneficial objective was given priority and the collective benefit was maximised. However, also the risk to the wearer had rapidly increased to the extent that it was not possible to predict its consequences (even negative).

Conversely, other people believe that the intended purpose of an object should be respected, in line with Substantialism’s theories, which attribute absolute value to an idea. According to this perspective, for example, a bin bag would be designed, tested and marketed to contain rubbish, not to protect healthcare workers from diseases (in this case the design principles and the tests will be different and stricter). Thus, these people tend towards a minimisation of the risk, but, at the same time, their precautional approach hinders the possible benefit underlying the other “less safe” alternatives. In this regard, it is necessary to recall the philosophy of Hans Jonas who, faced with an indeterminate and potential risk, the consequences of which cannot be estimated, introduced the imperative of responsibility in defence of future generations and based on the precautionary principle. His "heuristics of fear" implies foresight and ability to predict and adequately assess the consequences of collective activities in contemporary societies. Such principle implies to “act so that the effects of your action are compatible with the permanence of genuine life” and, in our present choices, to “include the future wholeness of Man among the objects of your (our) will” [ 30 ].

Consequently, this responsibility goes beyond the personal one of engineers, as it also includes the responsibility they partially assume if and when they do not limit reckless or inadequately considered actions, guided by the above-mentioned utilitarian approach. In fact, the compliance with international standards and the consequent CE marking does not only guarantee the quality, safety, efficiency, and efficacy of a product, but also the protection of manufacturers and users. In fact, as afore-mentioned from a general point of view, those to be blamed for the possible failure of bin bags used as PPE for the prevention of COVID-19 and the consequent infection and potential death of the healthcare workers using them are not the manufacturers of such items, but whomever decided to use this amateur substitution to other certified means of protection, and, to a certain extent, the biomedical engineers who did not respect their duty to identify the limits regarding the unintended uses.

Overall, two theoretical orientations are at the basis of these dichotomous approaches. However, in order to better frame them, it is necessary to analyse them in view of the extraordinary condition of necessity begotten by the pandemic. The dilemma revolves around the "intended use", or rather the purpose for which something (e.g., PPE or medical devices) was originally designed for: on the one hand there are those who assert that, in conditions of necessity, the contingent purpose, i.e., the social functionality that overcomes the intended use, ought to be preferred despite being “off label”. In fact, they firmly believe that it is preferable to maximize the current benefit while assuming an undefined risk. Although it is not easy to relate this trend to a specific current of thought, it certainly shares some points in common with utilitarianism, starting from the Benthamian one [ 31 ], if not with pragmatism (e.g., Dewey [ 32 ]).

On the other hand, the position of those who consider the “intended use” or rather the intrinsic purpose for which the product was manufactured tested and marketed, a priority, would seem evocative of Aristotelian substantialism or eschatology. In this case, the risks are limited by compliance with the law and the relevant standards, which also guarantee the achievement of the benefits. The refusal of this immediately relieves the manufacturer and the regulator from any responsibility related to the misuse of the object, leaving every possible and unforeseeable risk to the individual.

Here lies the crux of the problem. Using what is available and certified, albeit designed for a different “intended use”, seems more “reasonable” than not protecting oneself to everyone. However, we cannot refrain from asking ourselves the following questions: what is the limit within which it is possible to say, “better than nothing”? To what extent can science and policymakers put people's lives at risk in order to have a prompt, but probably unsafe answer in the wake of the "better than nothing" principle?

Regulatory frameworks and standards should be reviewed in this regard.

Beyond the DIY solutions, low-quality outputs have been affecting scientific production too. In fact, the high demand for information caused an acceleration in reporting scientific results, with many journals being overwhelmed with unprecedented numbers of papers, which challenged the capability of editors and reviewers to scrutinise articles.

The unprecedented high number of retracted papers can be a proxy for the high number of low-quality research on COVID-19. For this reason, a rapid search for papers regarding COVID-19 or SARS-CoV-2, and the previous epidemic/pandemics (i.e., SARS, MERS, Swine Flu) as a comparison, was performed both on OvidSP and the Retraction Watch Database. As regards COVID-19 publications, there were 124 retracted papers out of 264,530, i.e., 4.68 per 10,000 papers (compared to 1.16 per 10,000 papers concerning the previous pandemics/epidemics). Although this proxy is to be taken with a grain of salt, it should be a wake-up call for further investigations. Similar levels of confusion could be observed also among scientists and experts invited by media to interpret available scientific evidence and technical guidance.

The above-mentioned issues contributed to beget and feed an infodemic , defined by the United Nation as “an over-abundance of information—some accurate and some not—that makes it hard for people to find trustworthy sources and reliable guidance when they need it” [ 33 ], which is inducing an unprecedented need for responsible silence too.

Inadequacy of regulatory frameworks and norms

The pandemic creates a generalized condition of resource limited settings (RLSs), i.e., environments lacking means, specific knowledge, specialized personnel, medical devices and drugs within inappropriate medical locations. While this condition was already familiar to low- and middle-income countries, COVID-19 has overwhelmingly created RLS conditions in high-income countries, such as Europe, the USA and Japan, for the first time since World War II. This demonstrates how regulatory frameworks for medical devices and PPE are inadequate to RLS conditions. In fact, these regulations usually take into consideration standards that are too stringent and generic, proving impossible to fulfil in RLSs and, in times of the pandemic, difficult to adhere to universally. For example, the numerous tests and verifications required to assess the conformity of market respiratory protective equipment or eye protection equipment for healthcare purposes wasted time. One reason for this is that these standards are influenced by big manufacturers interested in having the largest market share. As a result, commercial standards for PPEs require testing in conditions that are not relevant for hospital workers (e.g., high temperature typical of heavy metals industry). Hence, international standards and norms followed the principle of generalism, losing universality and creating unnecessary burdens for small manufacturers [ 34 ]. In this regard, the WHO has published, for the first time, technical guidance on PPE specifically relevant for healthcare settings [ 35 , 36 ]. Differently to ISO standards for masks and respirators, the WHO guidance focuses on essential parameters, such as filtering capability, fit and breathability for masks.

Recalling the two aforementioned ethical perspectives, considering existing medical device regulation too generic to be universal, does not mean adhering to the utilitarian-pragmatic current tout court. There may be a contextualised response, regulated on the basis of tests, complying with flexible standards, or rather standards that are purposely designed to take into account different niche conditions. However, the use of any object must be certified and not random, and subject to tests relating to its specific intended use. Only in this way, people and their rights can be safeguarded, and science can prove to think deeply, act consciously and remain silent, when appropriate.

Contextualism is the basis of situational ethics [ 37 ], which seems to be the most adequate response to the specific needs of everyone and be able to face emergencies. In fact, it starts from the particular situation and tries to find universalizable answers, applying a heuristic and inductive method, progressive negotiations and interdisciplinary exchanges.

Conclusions

One year after the start of the pandemic, the need for ethic guidance is still tangible in everyday circumstances and essential during crisis or in RLSs. Respecting fundamental ethical principles while negotiating among different criteria (hospitalization demands vs available ICUs, generalism vs particularism, action vs responsible-action) requires clear guidance, deep knowledge, and peer-to-peer discussion among experts of different disciplines. The need for extreme specialization should never result in the fragmentation of knowledge.

Exactly a century, i.e., the Short Twentieth Century , separates COVID-19 from the last pandemic, the so-called "Spanish Flu", which flagellated Europe in 1918–1919. According to Hobsbawm, “ no period in history has been more penetrated by and more dependent on the natural sciences ” and “ yet no period, since Galileo's recantation, has been less at ease with it ”. This chasm between scientists and the general public is still open and, in some cases, fomented by populisms, which leverage on people's fears evoking war atmospheres, which have nothing to do with the catastrophic failure of many national healthcare systems’ response to this crisis. After a century, the dependence of medicine on biomedical science and engineering is evident, while their contribution to the definition of effective policies and norms is still negligible. Finally, the Cartesian fragmentation of knowledge, or rather “thinking in silos”, has persisted across the last century, calling for the creation of more fora where multidisciplinary discussions can be promoted. Three main needs emerged clearly: the need for responsible thinking, the need for responsible action and the need for responsible silence, when required and appropriate [ 38 ].

Availability of data and materials

Not applicable.

https://www.ft.com/content/491a4af4-66e7-11ea-a3c9-1fe6fedcca75 .

https://www.nytimes.com/2020/03/20/us/politics/trump-coronavirus-supplies.html .

https://www.lavanguardia.com/vida/20200411/48413026980/seat-detiene-produccion-respidradores-emergencia-descongestion-uci.html .

https://uk-mobile-reuters-com.cdn.ampproject.org/c/s/uk.mobile.reuters.com/article/amp/idUKKCN21U0UI .

https://www.franceinter.fr/coronavirus-8-500-respirateurs-produits-pour-rien .

https://www.bbc.co.uk/news/health-52145140 .

https://www.telegraph.co.uk/news/2020/04/08/exclusive-three-nurses-forced-wear-bin-bags-ppe-shortage-test/ .

Abbreviations

World Health Organization

Biomedical Engineering

European Alliance for Medical and Biological Engineering & Science

International Union for Physical and Engineering Sciences in Medicine

Intensive Care Unit

Società Italiana di Anestesia Analgesia Rianimazione e Terapia Intensiva

National Health Service

Personal Protective Equipment

French S. Between factualism and substantialism: Structuralism as a third way. Int J Philos Stud. 2018;26(5):701–21.

Article   Google Scholar  

Morin E. From the concept of system to the paradigm of complexity. J Soc Evolut Syst. 1992;15(4):371–85.

Department for Health of Ireland. Ethical framework for decision-making in a pandemic. https://assets.gov.ie/72072/989943ddd0774e7aa1c01cc9d428b159.pdf (2020). Accessed 19 June 2020.

The Nuffield Council of Bioethics. Responding to the COVID-19 Pandemic: Ethical Considerations. https://www.nuffieldbioethics.org/news/responding-to-the-covid-19-pandemic-ethical-considerations (2020). Accessed 19 June 2020.

Comité De Bioética De España. Informe Del Comité De Bioética De España Sobre Los Aspectos Bioéticos De La Priorización De Recursos Sanitarios En El Contexto De La Crisis Del Coronavirus. http://assets.comitedebioetica.es/files/documentacion/Informe CBE-Priorizacion de recursos sanitarios-coronavirus CBE.pdf (2020). Accessed 19 June 2020.

Deutscher Ethikrat. Solidarity and responsibility during the coronavirus crisis. https://www.ethikrat.org/en/press-releases/2020/solidarity-and-responsibility-during-the-coronavirus-crisis/ (2020). Accessed 19 June 2020.

UNESCO. Ethics in Research in Times of Pandemic COVID-19. https://en.unesco.org/news/ethics-research-times-pandemic-covid-19 (2020). Accessed 19 June 2020.

UNESCO. Statement on COVID-19: ethical considerations from a global perspective. https://unesdoc.unesco.org/ark:/48223/pf0000373115 (2020). Accessed 19 June 2020.

WHO. Ethical standards for research during public health emergencies: distilling existing guidance to support COVID-19 R&D. https://www.who.int/blueprint/priority-diseases/key-action/liverecovery-save-of-ethical-standards-for-research-during-public-health-emergencies.pdf?ua=1 (2020). Accessed 19 June 2020.

Berlinger N, Wynia M, Powell T, Hester DM, Milliken A, Fabi R, et al. Ethical framework for health care institutions responding to novel Coronavirus SARS-CoV-2 (COVID-19) guidelines for institutional ethics services responding to COVID-19. Safeguarding communities, guiding practice, 2; 2020.

Meyfroidt G, Vlieghe E, Biston P, De Decker K, Wittebole X, Collin V, et al. Ethical principles concerning proportionality of critical care during the COVID-19 pandemic: advice by the Belgian Society of IC medicine. Retrieved 2 April 2020. https://www.hartcentrumhasselt.be/professioneel (2020).

Emanuel EJ, Persad G, Upshur R, Thome B, Parker M, Glickman A, et al. Fair allocation of scarce medical resources in the time of Covid-19. Waltham: Mass Medical Soc.; 2020.

Book   Google Scholar  

Smith MJ, Ahmad A, Arawi T, Dawson A, Emanuel EJ, Garani-Papadatos T, et al. Top five ethical lessons of COVID-19 that the world must learn. Wellcome Open Res. 2021;6:17.

Smith MJ, Upshur RE. Learning lessons from COVID-19 requires recognizing moral failures. J Bioeth Inq. 2020;17(4):563–6.

Smith MJ, Upshur RE, Emanuel EJ. Publication ethics during public health emergencies such as the COVID-19 pandemic. Am J Public Health. 2020;110(7):947–8.

WHO. Resources on ethics and COVID-19. https://www.who.int/ethics/topics/outbreaks-emergencies/covid-19/en/ (2020). Accessed 19 June 2020.

WHO. Addressing human rights as key to the COVID-19 response. https://www.who.int/publications-detail/addressing-human-rights-as-key-to-the-covid-19-response (2020). Accessed 19 June 2020.

Comitato nazionale per la bioetica. Covid 19: clinical decision-making in conditions of resource shortage and the "pandemic emergency triage criterion". http://bioetica.governo.it/media/3987/p136_2020_covid-19-la-decisione-clinica-in-condizioni-di-carenza-di-risorse-e-il-criterio-del-triage-in-emergenza-pandemica.pdf (2020). Accessed 19 June 2020.

WHO Preparedness. https://www.who.int/environmental_health_emergencies/preparedness/en/ (2020). Accessed 19 June 2020.

The only cure is solidarity from Jürgen Habermas. https://www.classlifestyle.com/news/42575/kura-e-vetme-eshte-solidariteti-nga-jrgen-habermas/eng (2020). Accessed 19 June 2020.

Vergano M, Bertolini G, Giannini A, Gristina GR, Livigni S, Mistraletti G, et al. Clinical ethics recommendations for the allocation of intensive care treatments in exceptional, resource-limited circumstances: the Italian perspective during the COVID-19 epidemic. Berlin: Springer; 2020.

Google Scholar  

Rosenbaum L. Facing COVID-19 in Italy—ethics, logistics, and therapeutics on the epidemic’s front line. N Engl J Med. 2020;382(20):1873–5.

Nicoli F, Gasparetto A. Italy in a time of emergency and scarce resources: the need for embedding ethical reflection in social and clinical settings. J Clin Ethics. 2020;31(1):92–4.

Ranney ML, Griffeth V, Jha AK. Critical supply shortages—the need for ventilators and personal protective equipment during the COVID-19 pandemic. N Engl J Med. 2020;382(18):e41.

Sokolov P, Ivanova J. 17th century political cartesianism and its opponents, or imaging the state from point fixe. Higher School of Economics Research Paper No. BRP, 8. 2012.

Jones D. Censorship: a world encyclopedia. London: Routledge; 2001.

Maasen S, Weingart P. Democratization of expertise? Exploring novel forms of scientific advice in political decision-making, vol. 24. Berlin: Springer; 2006.

Bennett B. Logically fallacious: the ultimate collection of over 300 logical fallacies (Academic Edition). EBookIt.com. 2017.

Regulation EU 2017/745 of the European Parliament and of The Council of 5 April 2017 on medical devices, amending Directive 2001/83/EC, Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and repealing Council Directives 90/385/EEC and 93/42/EEC (2017). In: The European parliament and the council of the European Union (Ed.).

Jonas H. The imperative of responsibility: In search of an ethics for the technological age. 1984.

Bentham J, Mill JS. Utilitarianism and other essays. London: Penguin; 2004.

Dewey J. The essential Dewey: pragmatism, education, democracy, vol. 1. Bloomington: Indiana University Press; 1998.

WHO. Novel coronavirus (2019-nCoV) Situation Report-13; 2020.

Pecchia L, Piaggio D, Maccaro A, Formisano C, Iadanza E. The inadequacy of regulatory frameworks in time of crisis and in low-resource settings: personal protective equipment and COVID-19. Health Technol. 2020;10:1375–83.

WHO. Rational use of personal protective equipment for COVID-19 and considerations during severe shortages: interim guidance, 23 December 2020. World Health Organization; 2020.

WHO. Technical specifications of personal protective equipment for COVID-19.

Fletcher JF. Situation ethics: the new morality. Louisville: Westminster John Knox Press; 1997.

Maccaro A, Piaggio D, Pagliara S, Pecchia L. The role of ethics in science: a systematic literature review from the first wave of COVID-19. Health Technol. 2021;3:1–9.

Download references

Acknowledgements

The authors would like to acknowledge Hardip Boparai (PhD student at the University of Warwick, UK) and Katy Stokes (PhD student at the University of Warwick, UK) for proofreading this manuscript and for their valuable suggestions.

DP and LP received support from the University of Warwick with two Warwick Impact Found Grants supported by the EPSRC Impact Accelerator Award (EP/K503848/1 and EP/R511808/1). AM’s Fellowship is supported by the WIRL COFUND – Marie Sklodowska Curie Actions, Institute of Advanced Study, University of Warwick (UK). DP, LP, and AM also received support from Health Global Research Priorities of the University of Warwick. By supporting the researchers, the funders indirectly supported all the parts of this study.

Author information

Alessia Maccaro and Davide Piaggio contributed equally to this work.

Authors and Affiliations

School of Engineering, University of Warwick, Coventry, CV47AL, UK

Alessia Maccaro, Davide Piaggio & Leandro Pecchia

Institute of Advanced Study, University of Warwick, Coventry, CV47AL, UK

Alessia Maccaro

Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Italy

Concetta Anna Dodaro

European Alliance of Medical and Biological Engineering and Science (EAMBES), Leuven, Belgium

Leandro Pecchia

IUPESM, York, UK

You can also search for this author in PubMed   Google Scholar

Contributions

Conceptualization (AM and LP), data curation (AM, DP), Funding acquisition (LP), investigation (AM, DP, CAD), Project administration (LP), Supervision (LP, CAD), Writing original draft (AM, DP), Writing review and editing (AM, DP, LP, CAD). All authors reviewed the manuscript and agreed with the final version.

Corresponding author

Correspondence to Davide Piaggio .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

The authors declare that they have no conflict of interest.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Maccaro, A., Piaggio, D., Dodaro, C.A. et al. Biomedical engineering and ethics: reflections on medical devices and PPE during the first wave of COVID-19. BMC Med Ethics 22 , 130 (2021). https://doi.org/10.1186/s12910-021-00697-1

Download citation

Received : 17 August 2020

Accepted : 12 July 2021

Published : 25 September 2021

DOI : https://doi.org/10.1186/s12910-021-00697-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

  • Biomedical engineering
  • Regulations
  • Responsibility

BMC Medical Ethics

ISSN: 1472-6939

recent research papers of biomedical engineering

Volume 24 Supplement 2

The 8th International Work-Conference on Bioinformatics and Biomedical Engineering - Editorial

  • Introduction
  • Open access
  • Published: 18 October 2023

Advances and challenges in Bioinformatics and Biomedical Engineering: IWBBIO 2020

  • Olga Valenzuela 1   na1 ,
  • Mario Cannataro 2   na1 ,
  • Irena Rusur 3   na1 ,
  • Jianxin Wang 4   na1 ,
  • Zhongming Zhao 5   na1 &
  • Ignacio Rojas 6   na1  

BMC Bioinformatics volume  24 , Article number:  361 ( 2023 ) Cite this article

945 Accesses

Metrics details

This Supplement issue, presents five research articles which are distributed, mainly due to the subject they address, from the 8th International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO 2020), which was held on line, during September, 30th–2nd October, 2020. These contributions have been chosen because of their quality and the importance of their findings. Those contributions were then invited to participate in this supplement for the following journals of BMC: BMC Bioinformatics and BMC Genomics. In the present Editorial in BMC journal, we summarize the contributions that provide a clear overview of the thematic areas covered by the IWBBIO conference, ranging from theoretical/review aspects to real-world applications of bioinformatic and biomedical engineering.

Introduction to the IWBBIO 2020 edition

International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO 2020) seeks to provide a discussion forum for scientists, engineers, educators and students about the latest ideas and realizations in the foundations, theory, models and applications in the field of bioinformatics and biomedical engineering.

It has been the first edition that has been held online, but due to the circumstances imposed by COVID-19, the safety and well-being of our participants were on top of the agenda. The conference was adapted to fulfill the regulations imposed by the competent authorities. The virtual presentations and the video conferences in real-time (using Zoom) were ultimately a great success and presented without major problems.

One of the main objectives of the conference is that research in the bioinformatic field can reach medical applications. The conference sought to focus on diverse fields to create multidisciplinary research integrating areas like biomedical engineering, computer since, mathematics, artificial intelligence, bioinformatics, statistics or biomedicine [ 1 ].

These ideas provided important advances to the scientific community in fields like genomics, next-generation sequencing, drug design and advanced pharmacology, biomedical modelling and e-health, among other.

The list of topics in the successive Call for Papers has also evolved, resulting in the following list for the present edition:

Computational proteomics . Analysis of protein-protein interactions, Protein structure modelling, Analysis of protein functionality, Quantitative proteomics and post translational modifications (PTMs), Clinical proteomics, Protein annotation, Data mining in proteomics.

Next generation sequencing and sequence analysis . De novo sequencing re-sequencing and assembly, Expression estimation, Alternative splicing discovery, Pathway Analysis, Chip-seq and RNA-Seq analysis, Metagenomics, SNPs prediction.

High performance in bioinformatics . Parallelization for biomedical analysis, Biomedical and biological databases, Data mining and biological text processing, Large scale biomedical data integration, Biological and medical ontologies, Novel architecture and technologies (GPU, P2P, Grid,etc.) for Bioinformatics.

Biomedicine . Biomedical Computing, Personalized medicine, Nanomedicine, Medical education, Collaborative medicine, Biomedical signal analysis, Biomedicine in industry and society, Electrotherapy and radiotherapy.

Biomedical engineering . Computer-assisted surgery, Therapeutic engineering, Interactive 3D modelling, Clinical engineering, Telemedicine, Biosensors and data acquisition, Intelligent instrumentation, Patient Monitoring, Biomedical robotics, Bio-nanotechnology, Genetic engineering.

Computational systems for modelling biological processes . Inference of biological networks, Machine learning in Bioinformatics, Classification for biomedical data, Microarray Data Analysis, Simulation and visualization of biological systems, Molecular evolution and phylogenetic modelling.

Healthcare and diseases . Computational support for clinical decisions, Image visualization and signal analysis, Disease control and diagnosis, Genome-phenome analysis, Biomarker identification, Drug design, Computational immunology.

e-Health . e-Health technology and devices, e-Health information processing, Telemedicine/e-Health application and services, Medical Image Processing, Video techniques for medical images, Integration of classical medicine and e-Health.

During IWBBIO 2020 several Special Sessions have been carried out. Special Sessions were a very useful tool in order to complement the regular program with new and emerging topics of particular interest for the participating community. Special Sessions that emphasize on multi-disciplinary and transversal aspects, as well as cutting-edge topics were especially encouraged and welcomed, and in this edition of IWBBIO 2020 were the following:

SS1.High-throughput Genomics: Bioinformatic Tools and Medical Applications.

Genomics is concerned with the sequencing and analysis of an organism’s genome. It is involved in the understanding of how every single gene can affect the entire genome. This goal is mainly afforded using the current, cost-effective, high throughput sequencing technologies.

Organizers: Prof. Dr. Cecilio Angulo, Prof. Dr. Juan Antonio Ortega,Prof. Dr. Luis Gonzalez

SS2. Evolving Towards Digital Twins in Healthcare (EDITH).

Digital Twins is a very promising technique, as well as an ongoing research topic, imported from the Industry domain in order to develop Personalized Healthcare around the behavior of either patients’ disease or users’ health profile. The objective of this session is to present and discuss the advances in this important topic, Digital Twins, in the generation of knowledge. We advocates that this session will proportionate an important meeting point among different and variate researchers.

Organizers: Prof. M. Gonzalo Claros, Dr. Javier Pérez Florido, Dr. Francisco M. Ortuño

SS3. Data Mining from UV/VIS/NIR Imaging and Spectrophotometry.

This special section provided discussion on novel development, implementation, and approaches in sensors, measurements, methods, evaluating software, and data mining focused on the spectral and color analysis. The topic should cover practical examples, strong results, and future visions.

Organizer: Dr. Jan Urban

SS4: Intelligent Instrumentation.

Instruments and devices are almost similar and used for different scientific evaluations. They have become intelligent with the advancement in technology and by taking the help of artificial intelligence. In our daily life, sensors are corporate in several devices and applications for a better life. Such sensors as the tactile sensors are included in the touch screens and the computers’ touch pads. The input of these sensors is from the environment that converted into an electrical signal for further processing in the sensor system. The sensor’s main role is to measure a specific quantity and create a signal for interpretation.

Organizer: Prof. Dr. Barney

SS5. Image Visualization and Signal Analysis.

Any signal that is transmitted from a biological or medical source can be referred to as a biosignal. On the other hand, medical imaging is the technique and the process of creating visual representations of the inside of the body for clinical analyzes and medical interventions as well as the visual representation of the function of some organs or tissues (physiology). The medical images also create a database of normal anatomy and physiology to identify anomalies. Although imaging of harvested organs and tissues can be done for medical reasons, such procedures are generally considered part of the pathology rather than medical images.

Organizers: Prof. Dr. L.Wang

SS6. Analysis of Protein-protein Interactions.

Protein-protein interactions (PPI) are related to the association of proteins and the study of these associations from the perspective of biochemistry, signal transduction and protein interaction networks. Interactions between proteins are important in many biological processes.

Organizers: Dr. Yang

SS7. Computational Approaches for Drug Design and Personalized Medicine.

With continuous advancements of biomedical instruments and the associated ability to collect diverse types of valuable biological data, numerous recent research studies have been focusing on how to best extract useful information from the ‘Big biomedical Data’ currently available.

Organizer: Prof. Dr. Hesham H. Ali

Contributions of this special issue

Those papers that were deemed particularly relevant, taking into account the evaluation and opinion of the reviewers and chairs, were then invited to participate in this supplement for the following BMC journals: BMC Bioinformatics and BMC Genomics.

The first paper authored by Dimitris Grigoriadis et al. [ 2 ], presented a novel Deep Learning-based method for effective removal of noisy CAGE signals. The distribution, abundance, and utilization of transcription start sites (TSS) within promoters is poorly understood. Cap Analysis of Gene Expression (CAGE) has become a popular protocol for gene expression profiling that quantifies the usage of TSS by detecting the 5’ end of capped RNA molecules. These results highlight the need for computational methods that can effectively remove the excessive amount of noise from CAGE samples, leading to accurate TSS annotation and quantification of promoter usage. Regardless of sample quality, there are a significant number of CAGE peaks that are not associated with transcription initiation events. Indeed, there are a growing number of studies in the literature suggesting that CAGE can also detect 5’-capping events that are byproducts of transcription.

This raises the need for computational methods that can accurately increase the signal-to-noise ratio in data from CAGE, leading to error-free annotation of transcription start sites (TSS) and quantification of regulatory region usage. In this paper, the authors presented DeepTSS, a novel computational method for processing CAGE samples that combines genomic signal processing (GSP), structural DNA features, evidence of evolutionary conservation, and raw DNA sequence with Deep Learning (DL) to provide predictions for a single nucleotide TSS with an unprecedented level of performance. DeepTSS outperformed existing algorithms on all benchmarks, achieving 98% precision and 96% sensitivity (accuracy 95.4%) on the protein-coding gene strategy, with 96.6% of positive predictions overlapping active chromatin and 98.3% and 92% colocalized with at least one transcription factor and H3K4me3 peak, respectively.

The article by Luca Cappelletti et al. [ 3 ] focused on the use of deep neural networks that can accurately predict active regulatory regions in specific cell lines. Noncoding DNA regions, which make up 98% of the total human genome, have historically been considered “junk DNA.” However, their importance is now recognised in the scientific community because noncoding cis-regulatory regions (CRRs) regulate the transcription of neighbouring genes and thus determine the spatio-temporal patterns of gene expression. The annotation and characterization of tissue-specific cis-regulatory elements (CREs) in non-coding DNA is an open challenge in computational genomics. Recent studies have shown that genetic variants occurring in CRRs are strongly correlated with pathogenicity or harmfulness.

Deep-learning techniques, have recently achieved cutting-edge results in this challenging computational task. In this study, the authors provided additional evidence that feed forward neural networks (FFNNs) can be trained on epigenetic data and one-dimensional convolutional neural networks (CNN) trained on DNA sequence data can successfully predict active regulatory regions in different cell lines. Authors showed that model selection using Bayesian optimization applied to both FFNN and CNN models can significantly improve the performance of deep neural networks by automatically finding models that best fit the data. Furthermore, they showed that techniques applied to balance active and inactive regulatory regions in the human genome in training and testing data can lead to overoptimistic or poor predictions. In this paper is recommended using actual unbalanced data that were not used to train the models to evaluate their generalization performance. The experimental results confirm that deep neural networks can accurately predict active regulatory regions in specific cell lines and that automatic model selection by Bayesian optimization improves the quality of the learner and that rebalancing of the data significantly affects the predictive performance of the models. Finally, the convolutional models achieve performance close to that of feed-forward models using epigenomic information.

Automatic annotation of protein functions is an important topic in the field of bioinformatics because protein annotation is inadequate due to the high cost and time-consuming manual procedures for function identification. To be useful, protein sequences must be annotated with functional properties such as Enzyme Commission (EC) numbers and Gene Ontology (GO) terms. The development of computational tools for automatic annotation that leverage the high-quality manual annotations already available in UniProtKB/SwissProt is an important research problem. In the paper of Bishnu Sarker et al. [ 4 ], the authors extend the GrAPFI (graph-based automatic protein function inference) method (Sarker et al. in BMC Bioinform 21, 2020; Sarker et al., in Proceedings of 7th international conference on complex networks and their applications, Cambridge, 2018), a Graph-based Automatic Protein Function Inference approach, to add to the GO annotation and rename it GrAPFI- GO.

The authors have presented a pruning technique based on semantic similarity to eliminate the outlier annotations and a hierarchical post-processing step to enrich the remaining annotations with term preprocessing. The authors proposed several types of similarity measures based on common neighbors in the graph. Moreover, the terms on GO are hierarchically arranged according to semantic parent–child relationships. Therefore, an efficient pruning and post-processing procedure that takes into account both semantic similarity and hierarchical relationships among GO terms has been presented. The authors produced experimental results comparing the GrAPFI-GO method with and without considering the similarity of common neighbors. They also tested the performance of GrAPFI- GO and other annotation tools for GO annotation on a benchmark of proteins with and without the proposed pruning and post-processing procedure. As conclusion, the authors highlight that the proposed semantic hierarchical post-processing can improve the performance of GrAPFI-GO and other annotation tools.

Bacterial typing is a technique used to distinguish between different strains within a species. Typing is an important tool in epidemiology as it helps to find sources of infection as they are transmitted, and it is also used for epidemiological surveillance. Typing methods such as pulsed field electrophoresis (PFGE) or multilocus sequence typing (MLST) are used in clinical practise. Unfortunately, the discriminatory power of these methods is not sufficient to distinguish closely related bacterial strains, and they should be combined with methods such as whole genome sequencing (WGS), which can even find single nucleotide variants. An alternative to these methods is mini-MLST, a rapid, inexpensive and robust method based on high-resolution enamel analysis. In the paper by Marketa Nykrynova et al. [ 5 ], the authors presented a pipeline for the detection of variable fragments in unmapped reads based on a modified hybrid assembly approach using data from a sequencing platform.

The authors demonstrated the ability to identify one variable fragment in de novo assembled scaffolds of 21 Escherichia coli genomes and three variable regions in scaffolds of 31 Klebsiella pneumoniae genomes. For each identified fragment, melting temperatures are calculated based on the nearest neighbor method to verify the discriminatory power of the mini-MLST. As the most important conclusion, the authors highlight that the identified variable regions can then be used in efficient laboratory methods for bacterial typing such as the mini-MLST with high discriminatory power and completely replace expensive methods such as the MLST. The results can and will be delivered in a shorter time, enabling immediate and rapid infection surveillance in clinical practice. A disadvantage of the proposed methods is the uncertainty in the data compiled de novo.

Conclusions and Acknowledgement

The articles presented in this special issue provide some recent progresses in to Bioinformatics and Biomedicine Engineering fields. As Guest editors, we would like to express our thankfulness to all the authors who contributed their high quality research to the achievement of this supplement. Also, we are very grateful to expert scientists that have actively collaborated with their recommendations and suggestions to review and improve these contributions. We specially thank to Mr. Omar El Bakry for his excellent and constant support with the publication and edition of this supplement. It has been an honor for us to participate in it.

We finally invite authors and readers of this supplement to submit their recent works to future editions of IWBBIO, which will be announced at https://iwbbio.ugr.es . We wish the readers can benefit from insights of these relevant papers, and contribute to these rapidly and dynamics growing areas.

Availability of data and materials

Not applicable.

Rojas I, Valenzuela O, Rojas F, Herrera LJ, Ortuño F. (eds.): Bioinformatics and Biomedical Engineering, IWBBIO2020. Lecture Notes in Computer Science, volume 12108; 2020.

Grigoriadis D, Perdikopanis N, Georgakilas GK, Hatzigeorgiou AG. DeepTSS: multi-branch convolutional neural network for transcription start site identification from CAGE data. BMC Bioinform. 2022. https://doi.org/10.1186/s12859-022-04945-y .

Article   Google Scholar  

Cappelletti L, Petrini A, Gliozzo J, Casiraghi E, Schubach M, Kircher M, Valentini G. Boosting tissue-specific prediction of active cis-regulatory regions through deep learning and Bayesian optimization techniques. BMC Bioinform. 2022. https://doi.org/10.1186/s12859-022-04582-5 .

Sarker B, Khare N, Devignes MD, Aridhi S. Improving automatic GO annotation with semantic similarity. BMC Bioinform. 2022. https://doi.org/10.1186/s12859-022-04958-7 .

Nykrynova M, Barton V, Bezdicek M, Lengerova M, Skutkova H. Identification of highly variable sequence fragments in unmapped reads for rapid bacterial genotyping. BMC Genomics. 2022. https://doi.org/10.1186/s12864-022-08550-4 .

Article   PubMed   PubMed Central   Google Scholar  

Download references

This research has been partially supported by the projects with reference PID2021-128317OB-I00 (Ministry of Spain) and P20-00163 (FEDER, Junta Andalucia).

Author information

Olga Valenzuela, Mario Cannataro, Irena Rusur, Jianxin Wang, Zhongming Zhao and Ignacio Rojas contributed equally to this work

Authors and Affiliations

Faculty of Sciences, Applied Mathematics, University of Granada, Avenida de Fuente Nueva, 18071, Granada, Spain

Olga Valenzuela

Data Analytics Research Center - Department of Medical and Surgical Sciences, University “Magna Græcia” of Catanzaro, Computer Engineering and Bioinformatics, Catanzaro, Italy

Mario Cannataro

Department of Informatic, University of Nantes, Rue de la Houssinière, BP 92208, 48565, Nantes, France

Irena Rusur

Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha, 410083, Hunan, China

Jianxin Wang

Center for Precision Health, School of Biomedical Informatics and School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, 77030, USA

Zhongming Zhao

Information Technology and Telecommunications Engineering, CITIC-UGR, University of Granada, Periodista Daniel Saucedo Aranda, 18071, Granada, Spain

Ignacio Rojas

You can also search for this author in PubMed   Google Scholar

Contributions

OV, and IR coordinated the selection of the best papers during the IWBBIO2020 conference, jointly analyzing both the opinions of the reviewers and the chairman of each of the sessions. IR coordinated the file transfer with the authors of each of the articles presented and with the reviewers. OV, MC, IR, JW, ZZ and IR analyzed each of the papers, along with the reviews received. All authors wrote, read, revised and approved the final manuscript.

Corresponding author

Correspondence to Ignacio Rojas .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Valenzuela, O., Cannataro, M., Rusur, I. et al. Advances and challenges in Bioinformatics and Biomedical Engineering: IWBBIO 2020. BMC Bioinformatics 24 (Suppl 2), 361 (2023). https://doi.org/10.1186/s12859-023-05448-0

Download citation

Published : 18 October 2023

DOI : https://doi.org/10.1186/s12859-023-05448-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

  • High performance in bioinformatics
  • Next generation sequencing and sequence analysis
  • Biomedicine
  • Biomedical engineering
  • Computational systems for modelling biological processes
  • Healthcare and diseases

BMC Bioinformatics

ISSN: 1471-2105

recent research papers of biomedical engineering

SciTechDaily

  • February 21, 2024 | Researchers Add New “Twist” to Classical Material Design
  • February 21, 2024 | Swarming Secrets: Bat ‘Nightclubs’ Hold the Key to Preventing Future Viral Outbreaks
  • February 21, 2024 | Scientists Uncover Key to Tastier and Healthier Potato Chips
  • February 21, 2024 | Eclipsing Silicon: The Emergence of Magnon-Based Computing Technologies
  • February 21, 2024 | Global Crisis: Groundwater Levels Are Sinking Ever Faster Around the World

Biomedical Engineering News

Schematic Illustration Showing the Role of GQDs

Health February 10, 2024

Next-Gen Tumor Treatment: Metal-Free Nanozyme From Graphene Quantum Dots

A team led by Professor Wang Hui at the Hefei Institutes of Physical Science, part of the Chinese Academy of Sciences, has developed a metal-free…

Metamaterial Orthopedic Implants

Health February 7, 2024

$557,000 Boost – Next-Generation Spinal Fusion Goes “Meta”

An engineer from Pitt has been awarded $557K in NIH funding to conduct the world’s first in vivo trials of metamaterial orthopedic implants. A civil…

Belly Fat Obesity Weight Loss

Health February 1, 2024

The Future of Weight Loss? New Vibrating Pill Developed by MIT Reduces Food Intake by 40%

Ingesting the device prior to eating may induce a sensation of satiety, deceiving the brain into believing that it’s time to stop eating. When you…

Brain Implant Art

Technology January 8, 2024

Not Science Fiction: Brain Implant May Enable Communication From Thoughts Alone

A prosthetic device deciphers signals from the brain’s speech center to predict what sound a person is trying to say. A team of neuroscientists, neurosurgeons,…

Handheld, Wireless System To Detect Alzheimer’s and Parkinson’s Biomarkers

Health December 31, 2023

A New Wireless, Handheld, Non-Invasive Device Successfully Detects Alzheimer’s and Parkinson’s Biomarkers

The next steps involve testing saliva and urine samples with the biosensor. An international team of scientists has created a handheld, non-invasive device that can…

Mesenchymal Cells

Biology December 26, 2023

UCLA Study Challenges Long-Held Beliefs in Bioengineering and Stem Cell Dynamics

UCLA stem cell scientists ID surprise genetic instructions for boosting protein secretion, with big implications for biotech and cell therapy. Mesenchymal stem cells, found in…

Synthetic Programmable Bacteria

Health December 21, 2023

The $1 Cure: How Programmable Bacteria Are Reshaping Cancer Therapy

Texas A&M University researchers are co-leading a $20 million project to develop a $1 cancer treatment. What if a single one-dollar dose could cure cancer?…

3D Laser Based Imaging

Technology December 15, 2023

Echoes of Innovation: Caltech’s 3D Leap in Laser Photoacoustic Imaging

Caltech’s improved photoacoustic imaging technology, PACTER, simplifies procedures, enables 3D imaging, and reduces operational complexity, marking a significant advancement in medical imaging. There are times…

Spinal Cord Stimulation Amputee

Health December 14, 2023

Revolutionizing Amputee Care: Spinal Cord Stimulation Alleviates Pain and Enhances Balance

Spinal cord stimulation by University of Pittsburgh researchers effectively reduces phantom limb pain and restores sensation in amputees, marking a significant advance in prosthetic technology….

Hip Pain Hip Fracture

Health December 9, 2023

Tiny Changes, Big Benefits: Simple Lifestyle Changes Are Vital for Preventing Hip Fractures

Implementing community-wide simple bone-strengthening strategies, not limited to high-risk individuals, may significantly reduce hip fractures, according to a recent Australian study. Simple, community-wide strategies to…

Unlocking the Secrets of Natural Materials

Science December 3, 2023

Silk Solutions: Unlocking the Secrets of Natural Materials To Weave a Better World

Professor Benedetto Marelli develops silk-based technologies with uses “from lab to fork,” including helping crops grow and preserving perishable foods. Growing up in Milan, Benedetto…

Quantum Biology Informs Better Gene Editing Tool

Technology November 11, 2023

Revolutionizing CRISPR: Quantum Biology and AI Merge to Enhance Genome Editing

Oak Ridge National Laboratory’s research in quantum biology and AI has significantly improved the efficiency of CRISPR Cas9 genome editing in microbes, aiding in renewable…

Biomedical Implant Device Art Concept

Biology November 9, 2023

Breathing Life Into Medicine: “Living Pharmacy” Implant Gets Oxygenating Boost

New device could improve the outcomes of cell-based therapies. Cell-based therapies show promise for drug delivery, replacing damaged tissues, harnessing the body’s own healing mechanisms,…

Heart Nanowire Technology

Health November 6, 2023

Nanowired Cardiac Organoids – New Technology Could Revolutionize Recovery After Heart Attack

Preclinical research indicates that nanowired cardiac organoids could one day repair hearts instead of just preventing further damage. Heart disease is a leading cause of…

No Diabetes Concept

Health October 30, 2023

Scientists Develop New “Game-Changing” Technology for Pregnant Women With Diabetes

Researchers at the University of East Anglia suggest that automated insulin delivery should be adopted for pregnant women with type 1 diabetes. The technology, referred…

Neuromusculoskeletal Hand Prosthesis

Technology October 14, 2023

Revolutionizing Prosthetics – Scientists Develop Bionic Hand That Merges With User’s Nervous and Skeletal Systems

A Swedish woman who lost her right hand due to a farming accident was implanted with a novel human-machine interface into her residual bone, nerves,…

DNA Genetics Technology Device Art Concept

Technology October 7, 2023

Rapid Disease Diagnosis: Bioengineering Breakthrough Boosts DNA Detection Sensitivity by 100x

Scientists discover that letting small amounts of DNA ‘dance’ can speed disease detection. UMass Amherst researchers have pushed forward the boundaries of biomedical engineering one…

AI Experimental Design Engineering Cell State

Technology October 6, 2023

MIT/Harvard Cellular Reprogramming Innovation Could Find Potent Cancer Killers and Regenerative Therapies

A More Effective Experimental Design for Engineering a Cell Into a New State MIT and Harvard researchers have developed a novel computational technique that can…

Five Cutting-edge Advances in Biomedical Engineering and Their Applications in Medicine

Alliance of 50 experts from 34 elite universities reveals pioneering engineering advance across five vital domains.

Published Date

Share this:, article content.

Bridging precision engineering and precision medicine to create personalized physiology avatars. Pursuing on-demand tissue and organ engineering for human health. Revolutionizing neuroscience by using AI to engineer advanced brain interface systems. Engineering the immune system for health and wellness. Designing and engineering genomes for organism repurposing and genomic perturbations. 

These are the five research areas where the field of biomedical engineering has the potential to achieve tremendous impact on the field of medicine, according to “ Grand Challenges at the Interface of Engineering and Medicine ,” a study published by a 50-person task force published in the latest issue of IEEE Open Journal of Engineering in Medicine and Biology. The paper is backed by the IEEE Engineering in Medicine and Biology Society. 

{/exp:typographee}

Shankar Subramaniam is the lead author of the taskforce, distinguished professor in the Shu Chien-Gene Lay Department of Bioengineering at the University of California San Diego. Watch his presentation on the five grand challenges at the interface of engineering and medicine. 

“These grand challenges offer unique opportunities that can transform the practice of engineering and medicine,” said Shankar Subramaniam, lead author of the taskforce, distinguished professor in the Shu Chien-Gene Lay Department of Bioengineering at the University of California San Diego. “Innovations in the form of multi-scale sensors and devices, creation of humanoid avatars and the development of exceptionally realistic predictive models driven by AI can radically change our lifestyles and response to pathologies. Institutions can revolutionize education in biomedical and engineering, training the greatest minds to engage in the most important problem of all times – human health.”

In addition to Subramaniam, the following faculty from the UC San Diego Shu Chien-Gene Lay Department of Bioengineering were part of the task force: Stephanie Fraley, associate professor, Prashant Mali, professor, Berhard Palsson, Y.C. Fung Endowed Professor in Bioengineering and professor of pediatrics, and Kun Zhang, professor and a former department chair.

The study provides a roadmap to pursue transformative research work that, over the next decade, is expected to transform the practice of medicine. The advances would impact a wide range of conditions and diseases, from cancer, to diabetes, to transplants, to prosthetics.

The five grand challenges facing biomedical engineering

Bridging precision engineering and precision medicine for personalized physiology avatars In an increasingly digital age, we have technologies that gather immense amounts of data on patients, which clinicians can add to or pull from. Making use of this data to develop accurate models of physiology, called “avatars” – which take into account multimodal measurements and comorbidities, concomitant medications, potential risks and costs – can bridge individual patient data to hyper-personalized care, diagnosis, risk prediction, and treatment. Advanced technologies, such as wearable sensors and digital twins, can provide the basis of a solution to this challenge.

The pursuit of on-demand tissue and organ engineering for human health Tissue engineering is entering a pivotal period in which developing tissues and organs on demand, either as permanent or temporary implants, is becoming a reality. To shepherd the growth of this modality, key advancements in stem cell engineering and manufacturing – along with ancillary technologies such as gene editing – are required. Other forms of stem cell tools, such as organ-on-a-chip technology, can soon be built using a patient’s own cells and can make personalized predictions and serve as “avatars.”

Revolutionizing neuroscience using artificial intelligence (AI) to engineer advanced brain-interface systems Using AI, we have the opportunity to analyze the various states of the brain through everyday situations and real-world functioning to noninvasively pinpoint pathological brain function. Creating technology that does this is a monumental task, but one that is increasingly possible. Brain prosthetics, which supplement, replace or augment functions, can relieve the disease burden caused neurological conditions. Additionally, AI modeling of brain anatomy, physiology, and behavior, along with the synthesis of neural organoids, can unravel the complexities of the brain and bring us closer to understanding and treating these diseases.

Engineering the immune system for health and wellness With a heightened understanding of the fundamental science governing the immune system, we can strategically make use of the immune system to redesign human cells as therapeutic and medically invaluable technologies. The application of immunotherapy in cancer treatment provides evidence of the integration of engineering principles with innovations in vaccines, genome, epigenome and protein engineering, along with advancements in nanomedicine technology, functional genomics and synthetic transcriptional control.

Designing and engineering genomes for organism repurposing and genomic perturbations Despite the rapid advances in genomics in the past few decades, there are obstacles remaining in our ability to engineer genomic DNA. Understanding the design principles of the human genome and its activity can help us create solutions to many different diseases that involve engineering new functionality into human cells, effectively leveraging the epigenome and transcriptome, and building new cell-based therapeutics. Beyond that, there are still major hurdles in gene delivery methods for in vivo gene engineering, in which we see biomedical engineering being a component to the solution to this problem.

“We are living in unprecedented times where the collision of engineering and medicine is creating entirely novel strategies for human health. The outcome of our task force, with the emergence of the major research and training opportunities is likely to reverberate in both worlds--engineering and medicine--for decades to come” said Michael Miller, Professor and Director of the Department of Biomedical Engineering at Johns Hopkins University, who served as a senior author on the manuscript.

You May Also Like

Muscle as a heart-health predictor, capturing creativity with computation for music ai, nanoengineering department at uc san diego receives $2.1m gift from the aiiso yufeng li family, stay in the know.

Keep up with all the latest from UC San Diego. Subscribe to the newsletter today.

You have been successfully subscribed to the UC San Diego Today Newsletter.

Campus & Community

Arts & culture, visual storytelling.

  • Media Resources & Contacts

Signup to get the latest UC San Diego newsletters delivered to your inbox.

Award-winning publication highlighting the distinction, prestige and global impact of UC San Diego.

Popular Searches: Covid-19   Ukraine   Campus & Community   Arts & Culture   Voices

U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings
  • Advanced Search
  • Journal List
  • Sensors (Basel)

Logo of sensors

The Application of Biomedical Engineering Techniques to the Diagnosis and Management of Tropical Diseases: A Review

Fatimah ibrahim.

1 Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia; E-Mails: moc.liamtoh@oiht_treblig (T.H.G.T.); moc.oohay@82_girat (T.F.)

2 Centre for Innovation in Medical Engineering (CIME), Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

Tzer Hwai Gilbert Thio

3 Faculty of Science, Technology, Engineering and Mathematics, INTI International University, 71800 Nilai, Negeri Sembilan, Malaysia

Tarig Faisal

4 Faculty-Electronics Engineering, Ruwais College, Higher Colleges of Technology, Ruwais, P.O Box 12389, UAE

Michael Neuman

5 Department of Biomedical Engineering, Michigan Technological University, Houghton, MI 49931, USA; E-Mail: ude.utm@namuenm

This paper reviews a number of biomedical engineering approaches to help aid in the detection and treatment of tropical diseases such as dengue, malaria, cholera, schistosomiasis, lymphatic filariasis, ebola, leprosy, leishmaniasis, and American trypanosomiasis (Chagas). Many different forms of non-invasive approaches such as ultrasound, echocardiography and electrocardiography, bioelectrical impedance, optical detection, simplified and rapid serological tests such as lab-on-chip and micro-/nano-fluidic platforms and medical support systems such as artificial intelligence clinical support systems are discussed. The paper also reviewed the novel clinical diagnosis and management systems using artificial intelligence and bioelectrical impedance techniques for dengue clinical applications.

1. Introduction

The original designation of certain diseases as being tropical can be dated back to the 1898 publication of Sir Patrick Mansonʼs Tropical Diseases: A Manual of the Diseases of Warm Climates [ 1 ]. Tropical diseases (TD) are diseases that are widespread in or unique to the tropical and subtropical regions. TDs are prevalent in hot and humid climates. TDs are caused by pathogenic agents such as bacteria, viruses, parasites or fungi and are most often transmitted through carriers or vectors such as insects and nematodes. These insects may carry a parasite, bacterium or virus that is transmitted via their bite which disperses the infectious agent by subcutaneous blood or saliva exchange in humans and animals. Examples of tropical diseases include malaria, tetanus, hepatitis, American trypanosomiasis (chagas), dengue, yellow fever, cholera, nipah virus, and many others. We will look at some of these where biomedical engineering approaches have contributed to their diagnosis and treatment.

Rapid, efficient and inexpensive diagnosis of TDs is vital for the effective treatment and quality management of diseases in tropical regions. Diagnosis of TDs consists of a wide range of methods including serological testing of pathogenic markers such as protein, antigen, and antibody (using the enzyme-linked immunosorbent assay (ELISA) technique and various test kits), x-rays and physical examination, as well as performing bacteria and fungi culture techniques. These methods typically require a sample of bodily fluid, such as blood, sputum, or urine and/or stool samples. However, the management and diagnosis of TDs face several challenges such as prolonged turnaround time for assessment of specimens, high cost, a controlled environment, highly trained personnel, and large blood or bodily fluid samples. Associated measurements are often invasive, and in order to be useful, diagnostic methods must be accurate, simple and affordable for the population for which they are intended. They must also provide a result in time to institute effective treatment and patient isolation when necessary. Current diagnostic methods for TDs are frequently faced with challenges such as inconsistency in specificity and sensitivity, a long turnaround time to receive results, and high cost or specialization (which requires either an advanced and expensive laboratory setup, or skilled technicians and clinicians, or both).

Recent research and studies in the past decade have proposed and introduced various biomedical engineering approaches that attempt to address the issues faced in the diagnosis of tropical diseases. This review paper focuses on current biomedical engineering approaches for the more prominent diseases of dengue, malaria, cholera, schistosomiasis, lymphatic filariasis, ebola, leprosy, leishmaniasis, and American trypanosomiasis (chagas). Many of these approaches can be extended to other diseases as well.

These diseases were chosen as examples due to their severity and endemic nature and the perceived limitations of the current diagnostic methodology. In recent years outbreaks of dengue have been escalating and spreading geographically throughout the world with high mortality rates when timely treatment is not administered [ 2 , 3 ]. The number of infections of malaria and schistosomiasis is in the range of hundreds of millions, with the annual mortality rates in the range of hundreds of thousands [ 4 , 5 ]. From 2009 to 2010, the number of cases and deaths due to cholera has increased by about 50% [ 6 ], while sporadic outbreaks of ebola have been noted to be extremely deadly, killing the patient within 2 days after acquiring the infection [ 7 ]. The disease of lymphatic filariasisis is wide spread, infecting 120 million people and disfiguring 40 million worldwide [ 8 ], while leprosy has an annual infection rate of 200,000 a year, and causes permanent damage to the skin, nerves, limbs and eyes if untreated [ 9 ]. Around the world, millions of people are infected with leishmaniasis and American trypanosomiasis (chagas), and millions more are at risk due to the endemic nature of these diseases [ 10 , 11 , 12 ].

Although the diseases are different, many of the biomedical engineering approaches are common for their diagnosis and treatment. Table 1 presents the relation between the diseases and biomedical engineering approaches for their Diagnosis and Treatment. In the following sections we review biomedical engineering approaches that are currently available for diagnosis and establishing prognosis of the tropical diseases listed in Table 1 . As seen in the table, a wider variety of biomedical engineering approaches have been applied to diseases such as dengue and malaria than some of the diseases on the right side of the table.

Relation between the Tropical Diseases and their Diagnosis and Treatment via Biomedical Engineering approaches.

One of the most rapidly spreading mosquito-borne viral diseases in the world is dengue. The incidence of dengue has increased in the last 50 years as a result of increasing geographic expansion to new countries and to rural areas [ 2 ]. Annually, the World Health Organization (WHO) estimates the occurrence of 50 million dengue infections occurring and approximately 2.5 billion people facing threats of dengue since they live in dengue endemic areas [ 13 ].

According to the WHO 1997 guidebook of Dengue Haemorrhagic Fever: Diagnosis, Treatment, Prevention and Control, 2nd ed. [ 13 ], some patients develop dengue fever (DF) in the early stage after a person is infected with the dengue virus and recover after the fever subsides, while other patients may progress on to develop dengue haemorrhagic fever (DHF).

Laboratory and clinical diagnosis are used to diagnose the dengue patient. The clinical diagnosis and severity of DHF were graded from grades I to IV based on the WHO guidebook recommendation in 1997 [ 13 ]. The WHO guidelines define Grade I as patients having a fever accompanied by nonspecific constitutional symptoms with the only haemorrhagic manifestation being a petechial rash. Grade II is defined as patients having a spontaneous bleeding from any site on their skin. Grades III and IV are known as dengue shock syndrome (DSS). Grade III is where a patient has circulatory failure manifested by rapid and weak pulse, narrowing of pulse pressure (20 mmHg or less) or hypotension, with the presence of cold clammy skin and restlessness. Grade IV (DSS) is defined as patients having profound shock [ 13 ].

With the WHO 1997 dengue severity classification guideline [ 13 ], many clinicians experienced several difficulties especially in the critical cases such as the DHF patients which may experience significant plasma leakage that may lead to haemorrhage and organ impairment. The decision to admit those patients to the hospital in order to monitor their plasma leakage is a great challenge due to the overlapping of present medical classification criteria for establishing the risk of dengue patients [ 14 , 15 ]. On the other hand, physicians cannot decide to admit all patients because this will have an impact on the cost and quality of health care due to the high incidence of dengue in the tropics. Thus, to overcome these difficulties and to assist clinicians in determining the severity of the infection and how patients should be treated, the WHO has improved and introduced the new [ 2 ] guidebook which defines the dengue patients’ classification according to a few levels of severity. This guideline has been approved in year 2010 and valid until year 2014. In this new guidebook, the dengue patients are no longer classified as DF and DHF, instead patients are divided into two simplified groups. They are either severe or non-severe dengue patients. The non-severe dengue patients are further classified into two subgroups: patients with warning signs and those without warning signs.

In the new WHO 2009 guidebook [ 2 ], the classification criteria for non-severe dengue without warning signs are fever and any two of the following: nausea/vomiting, rash, aches and pain, a positive tourniquet test, leucopoenia, and there may even be a combination of these warning signs. The warning signs for non-severe dengue with warning signs are: abdominal pain or tenderness, persistent vomiting, clinical fluid accumulation (pleural effusion/ascites), mucosal bleeding, lethargy, restlessness, liver enlargement (>2 cm), and an increase in hematocrit (HCT) concurrent with a rapid decrease in platelet count. Such patients with these warning signs require strict observation and medical intervention. Severe dengue infections are characterized by significant plasma leakage, severe bleeding, and severe involvement of organs such as the liver. This causes liver enzymes such as aspartate aminotransferase (AST), or the alanine aminotransferase (ALT) to be elevated with readings of ≥1000 units/L.

Laboratory diagnosis of dengue patients is used to detect the dengue virus. Currently, a definitive diagnosis of dengue infection can be made only in the laboratory, either through virus isolation, detection of viral antigen or ribonucleic acid (RNA) present in serum or other bodily fluid, detection of antibodies present in serum, or a combination of these techniques [ 2 , 16 ]. At the early stage of infection, virus isolation and the detection of nucleic acid or antigen is used to make a diagnosis of dengue infection; however at the end of the acute phase of infection serological methods are more suitable [ 2 ]. A number of commercially rapid test kits (or coated strips) have been developed by various companies such as Korea’s Standard Diagnostics, Biorad and Panbio. These kits are able to detect a combination of either antigen or antibodies [ 3 ].

Although rapid laboratory diagnosis is very important and highly desirable, of equal importance is the recovery rate of the patient and the patient’s quality of life after recovery. To date, there is no effective vaccine or antiviral drug for dengue [ 2 ]. Some of the dengue patients might recover spontaneously while others face critical plasma loss that can lead to fatality [ 13 ]. Serology tests are not able to diagnose the micro-vascular status (micro-vascular leakage or plasma leakage) of the patient which is one of the major pathophysiological changes during dengue infection. Nevertheless the fatality of dengue disease can be reduced by close monitoring of patients to detect the onset of plasma leakage and administer prompt intravenous fluid replacement [ 17 ].

2.1. Biomedical Engineering Approaches

In order to overcome some of the difficulties in conventional dengue infection diagnostic methods, a few biomedical engineering (BME) approaches were introduced by proposing non-invasive tools to diagnose and classify the disease severity. These included techniques such as ultrasound imaging, echocardiography, electrocardiography, plethysmography, laser Doppler velocimetry, bioelectrical impedance, and intelligent clinical decision support systems. However, many of these papers reviewed in the following sections were based on the WHO 1997 [ 13 ] dengue classification as the new WHO 2009 [ 2 ] classification was introduced in the year of 2010, after these papers have been published.

2.1.1. Ultrasound

The critical stage in dengue occurs when the capillary permeability increases which leads to plasma leakage and therefore loss of plasma volume. The presence of pleural effusion and ascites are often used to determine the degree of plasma leakage. These are clinically detectable through physical examination techniques such as auscultatory percussion, and imaging techniques such as chest radiography, and also abdominal and thoracic ultrasonic imaging. Accordingly, several studies have used ultrasound as an aid for diagnosing dengue disease [ 18 , 19 , 20 ].

In the study by Srikiatkhachorn et al. [ 18 ] ultrasound has been employed to delineate the locations and the timing of plasma leakage in DHF. In the study, one hundred fifty-eight suspected dengue cases classified as DF, DHF, or other Febrile Illness (OFI) based on serology and evidence of plasma leakage including hemoconcentration and pleural effusion, were investigated. Ultrasound examinations of the abdomen and right thorax of patients were performed to detect ascites, thickened gall bladder wall, and pleural effusions. The results indicated that the timing of the plasma leakage was around the time of defervescence. The Pleural effusion was the most common ultrasonographic sign of plasma leakage while the thickening of the gallbladder wall and ascites were not associated as much in determining the plasma leakage. Significantly, plasma leakage of 12 out of 17 DHF cases who did not meet the WHO criteria for hemoconcentration signs was detected by ultrasound. The study concluded that ultrasound imaging is a useful tool for detecting plasma leakage in dengue infection.

Another study by Venkata et al. [ 19 ] was conducted to determine the importance of the ultrasound to clinical and laboratory profiles in diagnosing DF or DHF and to determine the usefulness of ultrasound in predicting the severity of the disease. One hundred twenty-eight suspected dengue patients (40 serologically negative for dengue fever and 88 serologically positive cases) were studied. Results of 32 patients from the 88 cases who were examined on the second to third day and repeated on fifth to seventh day showed that 100% had gall bladder wall thickening and pericholecystic fluid. Follow-up ultrasound on the fifth to seventh day showed ascites in 53%, left pleural effusion in 22%, and pericardial effusion in 28%. The results of the remaining 56 patients who were examined on the fifth to seventh day of fever for the first time showed that 100% had gall bladder wall thickening, 96% had ascites, 87.5% had right pleural effusion, and 66% had left pleural effusion. Contrary to the previous study, this study reported that thickened gall bladder wall, pleural effusion, and ascites should strongly favour the diagnosis of dengue fever.

In a separate work by Setiawan et al. [ 20 ], a study was conducted to examine the relationship between the clinical severity of 148 DHF patients (73 grades I and II; 75 grades III and IV) and their sonographic findings. Ultrasonography results revealed that the main features presented with grades I and II DHF patients were hepatomegaly 49%, ascites 34%, gallbladder wall thickening 32%, and pleural effusions 30%. On the other hand, the main features detected in DHF patients grades III and IV groups were pleural effusions, ascites and gallbladder wall thickening 95%, pararenal and perirenal fluid collections 77%, hepatomegaly 56%, and pancreatic gland enlargement 44%. The study concluded that ultrasound may be useful for early prediction of the severity of DHF.

2.1.2. Echocardiography and Electrocardiography (ECG)

Echocardiography and electrocardiography (ECG) were utilized in several studies to assess cardiac function of dengue patients [ 21 , 22 , 23 ].

Acute shock in severe DHF cases may occur in parallel with accumulation of fluid in serous body spaces such as the pleural, peritoneal, and pericardial cavities [ 21 ]. Pelupessy et al. [ 21 ] investigated the implementation of echocardiography in diagnosing dengue patients since echocardiography is a very sensitive method for detecting any small quantity of pericardial effusion The study showed that, although no signs of pericardial effusion could be determined on physical examination of DHF patients associated with severe shock and through ECG and radiological procedures, echocardiogram results were able to clearly show a small amount of fluid. Thus this technique is only recommended for the application of acute shock dengue patients.

In 1998, Wali et al. [ 22 ] employed radionuclide ventriculography, echocardiography, and ECG to assess cardiac function of 17 DHF and DSS patients. The radionuclide ventriculography results revealed that, the mean left-ventricular ejection fraction was 41.69%. Seven patients had an ejection fraction of less than 40%. Global hypokinesia was detected in 70.59% of the patients. The echocardiography results showed that the mean ejection fraction was 47.06%. The mean ejection fraction of the 8 DSS patients was 39.63%. Five (67.5%) of those patients had an ejection fraction below 40%. Radionuclide ventriculography and echocardiography showed no abnormalities after 3 weeks of follow up for five patients who had ST and T changes in their electrocardiogram. The ejection fraction was more than 50% in these cases. Within 3 weeks, the Global hypokinesia also improved and ECG changes reverted back to normal. The study concluded that acute reversible cardiac insult may be noticed in DHF/DSS and could be responsible for hypotension/shock seen in some of these patients. It was recommended that further studies are carried out to establish the pathogenic mechanisms of cardiac dysfunction in patients with DHF and DSS.

In 1993, Yusoff et al. [ 23 ] performed echocardiograms and ECGs on 28 consecutive adult patients with a clinical diagnosis of dengue infection. Twenty-three dengue patients were serologically confirmed (22 DHF grades I and II; 1 DF). 87% of the serologically confirmed dengue patients had abnormal ECGs and/or echocardiograms. Of these, 65% had abnormal ECGs that consisted of conduction abnormalities, ST segment elevation, T wave inversion, and sinus bradycardia. Fifty-two percent had abnormal echocardiograms which showed pericardial effusion, abnormal systolic and diastolic functions, left ventricular dilatation, and tricuspid regurgitation. The authors declared that ECG and echocardiographic abnormalities are common during the acute phase of DHF. They recommended early detection of cardiac involvement as a way of identifying the more severe forms of dengue so that appropriate treatment can be initiated as early as possible.

2.1.3. Strain Gauge Plethysmography

Liquid metal (Mercury-in-silicon elastomer tube) strain gauge plethysmography has been used in various studies to assess the microvascular permeability in dengue patients [ 24 , 25 ]. Gamble et al. [ 24 ] investigated the possible use of age-related changes in microvascular permeability as a health indicator, and it was found that the value was highest in DSS young children. These findings indicated that children significant factor in the susceptibility of children to DSS using strain gauge plethysmography [ 24 ]. Both adults and children DSS patients were found to have higher vascular permeability than the healthy control data, and the value was highest in the young children. These findings indicated that children are more susceptible to develop hypovolaemic shock than adults in DHF and other conditions characterized by increased microvascular permeability.

Bethell et al. [ 25 ], on the other hand, investigated whether the underlying pathophysiology of DSS is distinct from the milder forms of the disease by assessing the microvascular permeability also using strain gauge plethysmography. Three groups were investigated: children with DSS, DHF without shock, and in healthy children. The mean coefficient based on the statistical analysis of microvascular permeability for the patients with dengue was 50% higher than in healthy control patients. However, there was no significant difference in the permeability between the two groups of patients with dengue, which suggests the same underlying pathophysiology. This study also demonstrated that increased microvascular leakage occurs in children with DHF, and is most pronounced in children with DSS. However, the time taken to conduct the measurement was long (40 min average), and the procedure of cuffing the patients may put the patients at risk by inducing more capillary leakage.

2.1.4. Laser Doppler Velocimetry

Laser Doppler velocimetry has been widely used for assessment of various physiological parameters, particularly involving blood perfusion and circulation [ 26 , 27 , 28 , 29 ]. The technique has also been recently applied to DHF patients to evaluate the microcirculation changes due to plasma leakage and increase of microvascular permeability [ 30 ]. The preliminary findings of this study indicated that there were significant differences between basal laser Doppler flux in normal healthy subjects and DHF patients. These results also implied that the technique has the potential as an indicator to microcirculatory changes in DHF patients. The study has not; however been able to conclusively differentiate the DHF severity stages using this technique.

2.1.5. Bioelectrical Impedance

Over the years, bioelectrical impedance analysis (BIA) has demonstrated its utility as a non-invasive method for measurement and diagnosis in several medical applications. BIA evaluates the human body composition such as mass distribution ( i.e. , body cell mass and extracellular mass), and water compartments ( i.e. , intracellular water, extracellular water) by passing a small current through it [ 31 ]. The majority of the impedance measurements use four electrodes to minimize electrode-skin impedance effects. A small high-frequency current (50 kHz) is passed between two of the electrodes while the voltage drop across the same area is then measured using the other pair of electrodes. The ratio of the voltage drop to the current determines the resistance and reactance (Xc) of the body segment being measured. This data can then estimate the extracellular water (ECW), intracellular water (ICW), fat free mass (FFM), and fat mass through recognized equations [ 32 ].

BIA and bioelectrical impedance spectroscopy (BIS) have been described as a potentially reliable method to assess clinically significant changes in extracellular and total body water in dengue patients [ 33 , 34 , 35 , 36 , 37 ]. It could be a useful tool as a proxy for formal dilution methods to assess fluid shifts. Pierson and Wang (1986) have also proposed that an elevation of the extracellular water (ECW)/intracellular water (ICW) ratio in dengue patients, determined by dilution techniques, is a sensitive but nonspecific marker for the presence of systemic disease, and total body impedance measurements are especially useful in this determination.

In dengue infections several approaches were followed for utilizing the BIA technique. The hydrational profile which reflects the distribution of body water between the intra and extracellular space in dengue patients was investigated [ 38 , 39 , 40 , 41 , 42 , 43 , 44 ].

Studies conducted by Klassen et al. [ 42 ], Ibrahim et al. [ 38 ], and Mazariegos et al. [ 44 ] have shown that BIA was sensitive in determining the hydrational profile in dengue patients. Klassen et al. [ 42 ] studied the changes in hydrational status during the acute phase of classical dengue fever in 9 adult patients. They studied the effects of the acute classical dengue fever on ECW, ICW, and total body water (TBW) by comparing conventional dilution techniques with the outcome variables from whole body impedance spectroscopy (BIS), extracellular fluid resistance (R ecf ), and intracellular fluid resistance (R icf ) [ 42 ]. Two groups were investigated: a reference group comprised of 15 subjects without acute or chronic illness and dengue patients. The dengue patients were investigated on admission with febrile presentation (DI), at discharge after the defervescence of the fever cycle at about five days post-admission (DII), and seven days thereafter (DIII). The results revealed that the total body water did not change during the course of the disease and was not different from that in normal healthy subjects. However, the ratios of ECW/TBW and ECW/ICW reflected that body water shifted from the intracellular to the extracellular compartment in patients from the acute phase to convalescence. This ratio was also higher in convalescent dengue patients (DIII) compared to the reference group. The results also showed an association between increasing ECW, from the acute phase of the disease to convalescence and decreasing the R ecf and the R ecf /R icf ratio. Moreover, the R ecf and R ecf /R icf values were higher in the acute phase (DI) of dengue fever compared to those of the non-dengue subjects. The study concluded that relative expansion of ECW during the course of the disease and in the convalescence stage as determined by measuring body impedance can be used to monitor the dengue fever progression.

Fang et al. [ 45 ] reported a biosensor platform for dengue fever detection from patient serum. Their sensor was based on a non-faradic process where an integrated sensor captures the dengue antibody selectively from the sample. A thin film of sol-gel derived barium strontium titanate (BST) was coated on the immunosensor surface, and then the surface was modified with an organic self-assembled monolayer (SAM). In addition, pre-inactivated dengue virus was indirectly immobilized on the surface to act as a sensing probe to capture the dengue antibody. Finally, the modified surface was based on a finger shaped electrode where the output impedance/current will change in correlation with the presence and concentrations of dengue antibody in the serum sample. The work was conducted at frequency ranges of 0.1 Hz to 1 MHz.

Ibrahim et al. [ 46 ] monitored and modelled the hemoglobin (Hb) status in dengue patients using the BIA parameters. The Hb status was used since it is directly related to the Hct status which can be used to determine the degree of microvascular permeability in dengue patients. The BIA was employed to construct a model for predicting Hb in dengue patients using the multivariate analysis technique. Eighty-three (47 males and 36 females) serologically confirmed DF and DHF patients were studied during their hospitalization. The data consisted of all the investigated parameters in BIA, patientsʼ symptoms, and demographic data. Four predictors: reactance, gender, weight, and vomiting were found to be the most significant parameters for predicting the Hb levels in dengue patients. The study concluded that the single frequency bio-impedance technique and Multiple-linear-regression analysis is insufficient to monitor Hb for dengue patients since this analysis only explains approximately 42% of the Hb variation. This model has been enhanced by utilizing a non-linear artificial neural network (ANN) and achieved 74% accuracy [ 47 ]. The graphical user interface for the model is shown in Figure 1 . The enhanced model is able to predict the Hb concentration which helps to determine the degree of microvascular permeability in dengue patients based on the above mentioned data.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g001.jpg

Non-invasive haemoglobin modelling of dengue patients using bioelectric impedance analysis and artificial neural networks.

Additional enhancement of the model was achieved by employing the nonlinear autoregressive moving average with exogenous input (NARMAX) approach [ 48 ]. Three different NARMAX model order selection criteria, namely FPE, AIC, and Lipschitz were evaluated. The results gave 88.40% prediction accuracy by using the Lipschitz number approach. The study concluded that the NARMAX model yields better accuracy compared to the autoregressive moving average with exogenous input (ARMAX) model which achieved 76.70% accuracy.

2.1.6. Dengue Clinical Decision Support Systems

A physician’s knowledge would be sufficient to diagnose diseases that are directly related or corroborate to the encyclopaedic aspects of medicine. However, the complexity of diseases such as dengue and the many overlapping levels of its severity has created many difficulties for the physician to predict the disease prognosis. Accordingly, there is a crucial need for a decision support system to assist the health care provider in understanding the disease and to plan for its treatment. This is especially true when providers with less training than a physician are caring for patients, a factor crucially important in the developing world. In dengue, several approaches have been followed to achieve this goal including the self-organized map, multilayer feed-forward neural networks (MLFFNN), and adaptive neuro-fuzzy inference system (ANFIS) techniques.

Self-Organizing Map (SOM)

The self-organizing map (SOM) is an unsupervised neural network that is considered as one of the most powerful aids for visualizing, analyzing, and understanding the complexity of high-dimensional data. It receives a number of different multivariable input samples, discovers significant relationships in these samples, and presents them into a two dimensional map. This map contains different data clusters (prototypes), each of them consisting of samples that have similar features. Similarly, relationships within the data and cluster structures can be visualized and interpreted. Therefore, the SOM can be considered as an exploratory data analysis tool for generating hypotheses on the relationships among the data. The SOM has been widely used in medical applications [ 49 , 50 , 51 , 52 , 53 ]. Typical SOM can be visualized by using the U-matrix and the component planes. The U-matrix visualizes the distances between map units which is used to show the cluster structure of the map. Normally it is colored and these colors represent the distance between map units [ 54 ]. Colors with high values in the U-matrix indicate a cluster border while colors with low value indicate the clusters themselves. The value of the color is presented in the color scale beside the U-matrix. On the other hand, the component planes present values of all variable in each U-matrix map unit.

In dengue infections, the SOM was employed to identify the non-invasive significant prognosis factors that can distinguish between dengue patients and healthy subjects and also distinguish between the male and female patients [ 47 ]. The study presented a new approach to determine the significant prognosis factors in dengue patients utilizing the SOM. This technique showed the significant factors that can differentiate between dengue patients and the healthy subjects. Three hundred twenty-nine samples (210 dengue patients and 119 healthy subjects) were used in the study. Each sample contained 35 predictors (17 BIA parameters, 18 symptoms/signs). Two SOMs were constructed as shown in Figure 2 . Each map contains the U-matrix (on the top left of the map) and the component planes that represent the value of the variable in the U-matrix. The bottom left of each map shows the labels which indicate the type of patient in each cluster of the U-matrix map (healthy subject (H) or dengue patients (D)).The first map was constructed based on the BIA parameters data (variables) while the second map utilized the symptoms and signs data (variables). By visualizing the U-matrix and investigating the similarity between the clusters in the U-matrix and the component planes, the correlations between the dengue patients and the prognosis factors form the symptoms and signs and BIA parameters can be defined. The results revealed that, the significant BIA prognosis factors for differentiating the dengue patients from the healthy subjects were reactance, ICW, ratio of ECW/ICW, and ratio of the extracellular mass to body cell mass. On the other hand, abdominal epigastic pain, petechiea rash, and bleeding tendency were the main signs and symptoms that were present in dengue patients.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g002.jpg

( a ) Visualization of the self organizing maps for the bioelectric impedance analysis parameters; ( b ) Symptoms, and signs data. Reproduced with permission [ 47 ].

Due to the limitations of the WHO 1997 classification criteria that have been used to classify the severity of dengue patients, Faisal et al. [ 55 ] employed clustering, the SOM’s technique to determine new criteria that may help classify dengue patients based on disease severity. This technique aims to apply the K-mean clustering technique to cluster the SOMʼs prototypes rather than clustering the data directly to enhance the data clustering. Generally, the implementation of this approach is performed in two stages: First, the SOM is trained to identify the prototypes of the dengue patients’ data, and second, the K-mean clustering technique is implemented to cluster those prototypes. As a result, three criteria were then defined to classify the level of risk in dengue patients. The results were validated by comparing them to some other dengue researchers’ findings as well as the WHO criteria [ 2 , 13 ]. By using this technique, important results were obtained: the new risk criteria classified 33% of the DF patients as high risk dengue patients. Those patients might not be hospitalized according to the WHO criteria since they were classified as DF patients. However, those patients were classified as high risk by using the new criteria and thereby they might be at risk and face death if they are not closely monitored to detect the onset of plasma leakage. Another significant result was that 65.5% and 57.7% of the patients who were classified by WHO as DHF I and DHF II, respectively, were classified by the new criteria as low risk dengue patients. Those patients need not be hospitalized since they are classified as low risk patients and therefore the savings on the cost of the hospital admissions can be substantial. This result agrees with other researchers’ findings [ 14 ] and the recent WHO guidelines which indicates there is a problem using the existing WHO classification due to the changes in the epidemiology of dengue, and there is a high potential for the clinicians’ decision to be based the levels of severity for classifying the patients [ 2 ].

Multilayer Feed-Forward Neural Networks (MLFFNN)

ANNs have been successfully applied to several problems in dengue disease. Ibrahim et al. [ 56 ] employed the multi-layer perceptron (MLP) network trained via the back-propagation (BP) algorithm to develop a prediction system for predicting the day of fever reduction in dengue patients due to the fact that the progression of the DHF patient to DSS occurs following this day. Ninety percent prediction accuracy was achieved by using this approach. The study concluded that since most of the dengue patients were sick during or around the time of fever reduction, this ANN might be very promising to assist clinicians in the early determination of prognosis and in prescribing the management plan for their patients.

The network architecture designed by Ibrahim et al. [ 56 ] has been used to develop a GUI as shown in Figure 3 . The user interface is comprised of a few dialog boxes and radio buttons that request patient information and input data for the prediction process. The patient information includes patient identification and gender. The radio buttons are used to simplify user selection of the symptoms and the signs presented by the dengue patients. The “predict the day of defervescence of fever” button gives the estimated day of fever reduction. The entire system was compiled to function as a standalone application that can be used in any computer environment.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g003.jpg

Graphic interface screen for the prediction of day of defervescene of fever in dengue patients.

Ibrahim et al. [ 39 ] introduced a novel approach to classify the risk in DHF patients by using the BIA technique. A total of 184 (97 males and 87 females) serologically confirmed dengue patients (DHF I–IV) were studied during their hospitalization. The relationship between gender and group with the biological electrical tissue conductivity BETC parameters was studied employing univariate analysis of variance. The experimental results showed that BETC, specifically the reactance, was a potentially useful tool in classifying the risk factor of DHF patients.

The work by Ibrahim et al. [ 39 ] has been extended to diagnose the risk in dengue infections using an artificial neural network technique [ 57 ]. The study employed MLFFNN for classifying the risk stages in dengue patients. In this study, the severity of risk in dengue disease was quantified using the dengue patients’ blood data based on threshold values obtained from other researcher’s findings and the WHO classification system [ 58 ]. Data comprises of 223 healthy subjects and 207 dengue patients were arranged randomly into the training and testing in the ratio of 70:30. The ANN was trained via the steepest descent back propagation with momentum algorithm method. The optimum network architecture was determined by optimizing the training parameters. The optimization criteria was the sum squared error (SSE) and total classification accuracy of the network. The total classification was subjected to a 25% error tolerance. After the optimal ANN structure was determined, it was pruned using a weight eliminating method to enhance the system performance. The overall classification accuracy was 96.27% with 95.88%, 96.83%, and 95.81% for high risk, low risk, and healthy groups, respectively.

Continuing research has taken the neural network architecture obtained from Ibrahim et al. [ 57 ]. and developed an automatic dengue risk classification. Using the Matlab software, a GUI has been developed for use in real clinical applications as shown in Figure 4 . The user interface is comprised of a few dialog boxes and radio buttons that request patient information and input data for the diagnosis process. The DIAGNOSE button is used to initiate MLFFNN calculation for each given input. The system produces one of three outcomes: healthy, low risk, or high risk; and the result is displayed in the bottom most dialog box.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g004.jpg

An Automatic dengue risk diagnostic system using artificial neural networks and bioelectric impedance analysis techniques.

The same technique and the same data were used by Faisal et al. [ 59 ] to develop a non-invasive intelligent technique for diagnosing the risk stages in dengue patients using clinical manifestations (signs and symptoms) and the BIA measurements. An accuracy of 70% with 0.121 sum squared error was achieved by this model. The study concluded that such a screening system can aid physicians in the diagnosis of the risk and the prognosis of dengue patients, but it will not be definitive.

Although the Multi-Layer Perceptron (MLP) trained via the back propagation (BP) algorithm has demonstrated significant good performance in classification and prediction applications compared to statistical analysis, it suffers from a slow convergence rate and often yields suboptimal solutions [ 60 , 61 ]. To overcome this drawback, many researchers have employed the Levenberg-Marquardt (LM) algorithm [ 62 ] or Scaled Conjugate Gradient (SCG) algorithm [ 36 ] for training the MLP since these methods provide faster convergence and better estimation results. Those algorithms have been successfully used in medical applications for classifying and diagnosing several diseases.

Faisal et al. [ 59 ] constructed the dengue patient diagnostic model using the LM and SCG algorithms. Systematic procedures involving training, testing, and validation were followed to construct the diagnostic model so that a higher performance of the diagnostic model can be achieved and the robustness of overall diagnostic models can be maintained. Precise tuning of the internal training algorithms’ parameters was performed to attain the optimal model. Three-layer network is used. The activation function in hidden layer’s neurons is hyperbolic tangent sigmoid while in the output layer’s neurons are sigmoid transfer function. The 5-fold Cross Validation (CV) technique is implemented. The data are divided into five sets; each set contains 101 samples (45 high risk patients, 56 low risk patients). Four sets (404 samples) were used for training and the remaining set was used for testing. The training process was repeated for five times, at each time one of the sets was used as testing set. The results for optimization of the MFNN trained via the Levenberg-Marquardt algorithm showed that the optimal model achieved an average diagnostic accuracy of 70.7% with 73% sensitivity, 74% specificity and a 0.02 average MSE. By implementing the scaled conjugate gradient algorithm, the optimal diagnostic model achieved an average diagnostic accuracy of 75% with 0.01957 average MSE.

Adaptive Neuro-Fuzzy Inference System (ANFIS)

Even with the success of ANNs in a decision support system, the use of computerized decision making systems in clinical medicine is rather difficult due to the uncertainty of naturally occurring diseases. In such a situation, fuzzy set theory appears as an appropriate tool for a decision making system since it deals with uncertainty by applying our knowledge and experience directly without any explicit mathematical models. Fuzzy logic describes human thinking and reasoning in a mathematical framework by using several rule bases (IF-THEN) that require a number of human experts to carefully define the rules. Even though the fuzzy logic has been successfully implemented, there are some basic aspects of it that are in need of better understanding. First, the need for a standard method for transforming human knowledge or experience into the rule base and database of a fuzzy inference system is noted. Second, there is a need for effective methods for tuning the membership functions [ 63 ]. Based on those needs, the Adaptive Neuro-Fuzzy Inference System (ANFIS) was proposed to serve as a basis for constructing a set of fuzzy if-then rules with appropriate membership functions to generate the stipulated input-output pairs [ 63 ].

Faisal et al. [ 64 ] utilized (ANFIS) to develop a dengue patient diagnostic model. The development of the model was carried out in two steps: defining the initial ANFIS model architecture and training of the defined model. Two approaches were followed to define the initial ANFIS model architecture. In the first approach, the number of membership functions in the inputs and the output were systematically varied and the effect of this variation in the model performance was investigated. In the second approach, a subtractive clustering algorithm was assigned to determine the initial ANFIS model by optimizing the number of membership functions and fuzzy rules. After the initial model structure was defined, it was trained so that the differences between the output obtained from the model and actual output are minimized. The hybrid learning algorithm was employed for this task. The results of the first approach showed that the highest overall accuracy of 80.19% with 71% sensitivity and 86% specificity was achieved. For the second approach, average diagnostic accuracy was 86.13% with 87.5% and 86.7 sensitivity and specificity, respectively.

The graphic user interface for the new ANFIS is shown in Figure 5 . The user interface is comprised of four parts: patient information such as patient's identification, symptoms and signs, the bioelectrical impedance data, and the dengue patient risk diagnostic which determines the level of risk in dengue patients.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g005.jpg

Dengue patient diagnostic system based on Adaptive Neuro-Fuzzy Inference System.

2.1.7. Dengue Fever Detection Using Microfluidic Lab-On-a-Disc (LOD) and Lab-On-a-Chip (LOC) Platforms

Microfluidic devices have been proposed as a solution to overcome the many problems caused by the conventional dengue testing systems such as the high cost, reagent consumption, long turnaround time and complex procedures. Lab-on-a-disc (LOD) and lab-on-a-chip (LOC) are the most popular microfluidic platforms that have been reported by many researchers as a portable diagnostic device for dengue detection [ 65 , 66 , 67 ].

Ibrahim et al. [ 68 ] and Yusoff et al. [ 69 ] have discussed the fundamentals of the microfluidic compact disc (CD) and its application as a platform for ELISA detection of dengue non-structural glycoprotein 1 (NS-1), and Ibrahim et al. [ 68 ] designed and fabricated a microfluidic CD to automatically perform an ELISA test for dengue detection.

On the other hand, lab-on-a-chip (LOC) has been proposed as a precise, rapid, and low cost platform for dengue detection [ 70 , 71 ]. Lee et al. [ 70 ] reported an integrated microfluidic platform that can detect the dengue virus by coating magnetic microbeads with antibody. A multi-way micropump shown in Figure 6 is used for moving the serum sample, reagents and other buffers inside the microfluidic net (channels and chambers). Lee et al. [ 72 ] claimed that the time required to perform one test is 30 min, which is only 1/8th of the time required to detect dengue using a conventional testing platform. Weng et al. [ 71 ] reported a new microfluidic LOC platform for dengue detection where a suction method (instead of pumping) is used to move the liquid inside the device. Polydimethylsiloxane (PDMS) has been used for the chip fabrication and the surfaces are pre-functionalized to minimized protein adsorption. The dengue testing process can be finished within 30 min and the chip can stay stable for one month if it is stored at 4 °C.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g006.jpg

The magnetic bead-based microfluidic chip measures 53 mm × 37 mm. Reproduced with permission [ 72 ].

2.1.8. Dengue Virus Detection Using Paper-Based-Diagnostic Platform

The lab-on-paper, or paper-based diagnostic platform is a recently developed low cost method of implementing point of care diagnosis. The paper-based devices are easy to use, low cost, disposable, and can be used for a wide range of biomedical diagnosis [ 73 , 74 , 75 ].

A medical patch developed by Martinez et al. [ 73 ] allows for the use of chromatography paper as a low cost, low volume, and portable bioassay. The paper is patterned with test areas which are doped with the required reagents, which change in color when it reacts with the intended analytes. Matthews et al. [ 74 ] further enhanced the paper-based platform by developing an object identification algorithm that is light enough for use on mobile phones. The application utilizes the built-in camera of mobile phones for image capturing of the medical patch developed by Martinez et al. [ 73 ]. Once the image is captured, a developed mobile application then performs the necessary image processing and determines the disease state.

In a novel study conducted by Lo et al. [ 75 ], dengue virus is detected on a paper-based device for the detection of dengue virus by first amplifying the nucleic acids via reverse transcription loop-mediated isothermal amplification (RT-LAMP), then analysing the results using a colorimetric assay on paper. The process of RT-LAMP is performed on Dengue serotype-2 ribonucleic acid (RNA) using conventional microwell assay technique. Once the RT-LAMP process is completed, the DNA product is then dropped on a waxed patterned 96-well paper where it is mixed with biotin-11-deoxyuridine, and then conjugated with streptavidin horseradish peroxidase. The 96-well paper is then washed and scanned to obtain the colorimetric results on a computer.The study concluded that the combination of RT-LAMP andpaper-based colorimetric approach reduces the process time, while requiring very little sample volume and is suitable for the point-of-care application.

2.2. Summary

The summaries of biomedical engineering techniques in dengue disease are listed in Table 2 , Table 3 and Table 4 . Table 2 summarizes the biomedical engineering techniques of ultrasound, echocardiography and electrocardiography (ECG), strain gauge plethysmography, laser Doppler velocimetry and bioelectrical impedance in dengue disease. Table 3 summarizes the biomedical engineering techniques in dengue clinical decision support systems. Table 4 summarizes the biomedical engineering techniques for dengue fever detection using microfluidic lab-on-a-disc (LOD), lab-on-a-chip (LOC), and paper-based platforms.

Summary of biomedical engineering techniques of ultrasound, echocardiography and electrocardiography (ECG), strain gauge plethysmography, laser Doppler velocimetry and bioelectrical impedance in dengue disease.

Summary of biomedical engineering techniques in in dengue clinical decision support systems.

Summary of biomedical engineering techniques of microfluidic lab-on-a-disc (LOD), lab-on-a-chip (LOC), and paper-based platforms in dengue disease.

Malaria is an infection that affects humans and some animals and is transmitted by infected female Anopheles mosquitoes. Fever, headache, and in some severe cases patients progressing to coma or death, are the common symptoms of this disease. Sub-Saharan Africa, Asia, and the Americas are the areas where malaria is most prevalent [ 4 ]. In general, five types of the plasmodium have the capability of causing human infection: P. falciparum , P. vivax , P. ovale , P. malariae , and P. Falciparum are mainly found in Africa and some parts of Asia and South America [ 76 ]. P. falciparum is the common cause of severe malaria cases that can lead to death. P. vivax is less fatal but can lead to serious anaemia in children. Two hundred sixteen million cases of malaria were reported in 2011, where 81% of those cases were found in Africa [ 4 ]. It is estimated that around 86% of these cases are children under 5 years of age. The two common methods for diagnosis of malaria are light microscopy of blood and the rapid diagnostic tests (RDTs) [ 77 ]. The main advantage of the light microscopy is the low cost in endemic areas. However, the need for a highly trained operator and the lack of portability are the main drawbacks. RDT testing kits are preferable in other areas but the cost of the RDT kits is high.

3.1. Biomedical Engineering Approaches

Several biomedical engineering approaches for malaria detection have been described including microfluidic systems, image processing, and bioelectrical properties of blood. The following sections review these approaches.

3.1.1. Image Processing

Many researchers have reported the auto-detection of malaria infection by utilizing image processing techniques on microscopy images [ 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 ]. This work is based on images of cultured malarial parasites that were grown under a controlled environment rather than in patients’ blood. Rao [ 85 ] utilized the stained images of the P. falciparum to analyse its life cycle. At a later study by Sio et al. [ 83 ], new software that automatically counts the malarial parasite has been reported. The software focused on the counting of P. falciparum in images where the regular blood components and other types of noise are not present. The algorithm applied is able to differentiate between infected and uninfected red blood cells, and successfully count the parasites from peripheral blood specimens. Dimension and colour of components were used for identification. Halim et al [ 80 ] performed template matching techniques to detect red blood cells (RBC). Gray scale images were processed using different image process techniques for parasite detection. The second developed method utilizes the colour co-occurrence array that analyzes pixel colour index and the indicated colour of the surrounding pixels. The various techniques produced results with accuracy of 80%–88% and a sensitivity of 92%–98%.

3.1.2. Microfluidics

The portability and the small volume of sample required to perform a test are the main advantages of the microfluidic platforms [ 86 ]. These features, along with the low cost of the microfluidic platform make it ideal to be used on site and in resource-limited areas.

Different immunochromatographic methods (dipstick) of malaria infection detection have been described [ 87 , 88 ]. These platforms are easy to use, portable, and can be used in the most challenging environments. Most are based on the detection of malaria antigens from a patient’s blood sample. The targeted antigen types include histidine-rich protein-2 (HRP-2), aldolase, and plasmodium lactate dehydrogenase (pLDH). The developed methods have the ability to differentiate between the main malaria types through the immunogenic differences in the proteins. Figure 7 shows an example of a microfluidic platform (dipstick) that utilizes immunochromatographic lateral flow to detect malarial proteins (antigen (Ag)) that are extracted from patient blood [ 89 ]. The disadvantage of the microfluidic platforms is their accuracy which is less than that from gold standard microscopy techniques [ 90 , 91 ]. As the disease severity declines, the concentration of malarial antigens quickly decreases. Therefore, these detection techniques are good for detection of recurrent infection and not for observing the response of the patient to treatment.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g007.jpg

Schematic of the lateral flow strip to diagnose Malaria. (top) Layout of the strip, (middle) Flushing agent is added to help flush parasitized blood along the strip, and (bottom) visible lines indicate presence of antigens in the parasitized blood. Reproduced with permission [ 89 ].

Many platforms were proposed using the flow cytometry approaches for malaria detection have been reported [ 92 , 93 ]. Saito-Ito et al. [ 93 ] presented a fast, high sensitivity, and low error diagnostic platform that is capable of detecting the P. falciparum from a sample that contains erythrocytes and stained parasites. The sensing principle was based on forward-angle light scattering from an argon laser, and green fluorescence was utilized in this process. The author claimed that the required time to perform a full test is 2–3 min including sample preparation. Jiménez-Díaz et al. [ 94 ] proposed a new flow cytometric platform for malaria detection based on observing the differences of infected erythrocytes stained auto-fluorescence and DNA content with that of healthy cells. The author claimed that the proposed platform is rapid, simple to use, sensitive, and can accurately detect malarial pathogens.

3.1.3. Paper-Based Microfluidic Cartridge

A method of paper-based microfluidics has been reported to prepare stained malaria parasites for detection using traditional optical microscopy [ 95 ]. Horning et al. [ 95 ] presented a paper microfluidic cartridge for automated staining of malaria parasites. The cartridge is similar to a dipstick, but replaces the cellulose strip with paper. Blood is dropped onto a piece of dyeing hydrophilic paper which stains the malaria parasites. The blood is then channeled by capillary forces through the paper into an optically transparent slanted chamber which gradually get thinner. The slanted chamber produces a thick smear near the paper, and a thin smear towards the end of the chamber. The cartridge can then be examined using traditional optical microscopy. The authors claimed that the device is easy to use, fast, low cost, has good optical properties, and is suitable for automated microscopy. A comparison with the standard Giemsa smear technique has shown that the cartridge produces smears equal to that of a blood smear as prepared on a microscope slide by an expert microscopist.

3.1.4. Microarray Chip

Recently, Jin et al. [ 96 ] proposed a microwell array chip that enables the analysis of single live cells. The proposed system is capable to analyze more than 234,000 individual cells rapidly and efficiently.

Expanding on the work of Jin et al. [ 96 ], Yatsushiro et al. [ 97 ] reported a novel high-throughput screening and analysis system for malaria infection using microarray chip with a laser scanner ( Figure 8 ). They utilized a polystyrene chip that contains 20,944 micro-chambers fabricated using molding techniques. The author claimed that the system is 10 to 100 times more sensitive than the conventional light microscopy diagnosis method. This system requires 15 min to detect malaria parasites in erythrocytes extracted from centrifuged blood.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g008.jpg

Schematic diagram of the process for detection of malaria-infected erythrocytes on a cell microarray chip. ( a ) Erythrocytes stained with a nuclei-specific fluorescent dye, SYTO 59, for the staining of malaria nuclei dispersed on a cell microarray chip using a pipette, which led to the formation of a monolayer of erythrocytes in the microchambers; ( b ) Malaria-infected erythrocytes were detected using a microarray scanner with a confocal fluorescence laser by monitoring fluorescence-positive erythrocytes; ( c ) The target malaria-infected erythrocytes were analyzed quantitatively at the single-cell level. Reproduced with permission [ 97 ]—open access.

3.1.5. Dielectrophoresis (DEP)

Dielectrophoresis (DEP) technology is utilized by many researchers for cell manipulating and characterization [ 98 ]. DEP is the phenomenon that describes the motion of polarizable particles in a non-uniform electric field. It allows for the recognition of differentiated populations of particles based on their relative polarisibility. Separation between the living and dead cells is an example of application of DEP; taking advantage of the unique electrical properties of each bioparticle [ 99 ]. Aceti et al. [ 100 ] found that the dielectric properties of erythrocytes infected with the malarial parasite P. falciparum are different from normal erythrocytes. The electrical conductivity of the erythrocyte membrane increases sharply when it is infected with Malaria parasites. This sharp increase was due to changes in the composition, morphology and permeability of the erythrocyte membrane. Therefore, Gascoyne et al. [ 101 ] utilized the DEP technique to separate the malarial infected cells from the healthy cells by dielectrophoretic manipulation in a non-uniform electric field. Only a few microliters of blood are required to detect malaria on the DEP platform.

3.1.6. Bioelectrical Properties

Lonappan et al. [ 102 ] studied the microwave characteristics of malaria and normal blood samples using a cavity perturbation technique. This method has many advantages such as its simplicity, rapid performance, reliable results and the small sample volume required to run the test. A significant difference of the conductivity of malaria blood samples when compared with the normal healthy samples was seen at frequencies of 2 to 3 GHz. This measurement is a new novel in vitro method of diagnosing malaria at its onset using microwaves, and it will allow precautions to be taken early in the disease course such as proper preventative drugs, which will improve disease prognosis [ 102 ].

Wilson et al. [ 103 ] developed an integrated system with dark-field reflection-mode and cross polarization microscopy for the detection of hemozoin in fresh blood samples. Hemozoin is an iron-containing pigment resulting from the breakdown of haemoglobin, found in the malaria parasite. The presence of hemozoin results in a different light scattering effect when compared to healthy RBCs. The result shows that incorporating both methods doubles the contrast when compared to the individual techniques [ 103 ].

3.2. Summary

The summary of biomedical engineering techniques in malaria disease is listed in Table 5 .

Summary of biomedical engineering techniques in malaria disease.

Cholera is an acute intestinal infection as a result of ingestion of food contaminated with the vibrio cholerae bacteria. This infection has a short period of incubation from two to three days. If treatment is not administered immediately, it causes an enterotoxin that leads to vomiting and excessive flowing diarrhea that results in extensive dehydration. According to the WHO, the incidence of cholera has been growing, with approximately 317,000 cases and 7500 deaths in 2010, demonstrating an increase of 43% in the number of cases and 52% in the number of deaths compared to 2009 [ 6 ]. In areas where cholera is endemic, the mortality rate has increased for children and pregnant women. Two types of vaccines are available to protect against this disease, and it is easily treated by intravenous rehydration [ 104 ].

Although standard culture methods can be employed for detection of v. Cholera , limitations including low accuracy, the requirement of expertise and well equipped laboratory, and lengthy time for diagnosis make it impractical in diagnosis of individual cases. Dark field microscopy and a dipstick assay can be used to achieve a fast identification of v. Cholerae . These assays can yield results that differ with the definitive culturing methods, therefore fluorescent monoclonal antibody and PCR-based techniques have been developed to improve sensitivity and specificity of the assays [ 105 ].

4.1. Biomedical Engineering Approaches

A few biomedical approaches have been investigated to improve the detection of cholera toxin (CT).

4.1.1. Microfluidics

Bunyakul et al. [ 106 ] developed a microfluidic platform incorporating fluorescence and electrochemical detection techniques for the detection of cholera toxin subunit B (CTB). The microfluidic platform was fabricated using soft lithography on polydimethylsiloxane (PDMS). The sample with CTB concentrations of up to 1000 ng/mL is injected into the platform towards a capture zone withCTB-antibodies and ganglioside GM 1 receptor immobilized onto magnetic beads. The beads are then immobilized using an external magnet for washing and then transferred to a detection zone for either fluorescence detection, or electrochemical detection. The implementation of the microfluidic platform allows for limits of detection of 6.6 and 1.0 ng·mL −1 respectively for the fluorescence and electrochemical techniques.

4.1.2. Impedance Techniques

Due to the low levels of lethal dose (LD) in humans [ 107 ], there is a high demand for detection techniques capable to discriminate very low concentrations of the toxin. Labib et al. [ 108 ] developed an immunosensor assay for the detection of CTB using a capacitive method. The sensor consists of a gold electrode coated with monoclonal antibodies against CTB (anti-CTB) inserted into a capacitive flow cell. Potential pulses of 50 mV at 50 Hz are transmitted through the capacitive flow cell, and the transient currents evoked between the electrode and the capacitive flow cell are recorded and used to calculate the capacitance of the liquid. The capacitance value recorded is found to decrease when CTB binds with the anti-CTB coating on the electrode.

4.2. Summary

The summary of biomedical engineering techniques in cholera disease is listed in Table 6 .

Summary of biomedical engineering techniques in cholera disease.

5. Schistosomiasis

Schistosomiasis is a type of parasitic infectious disease affecting at least 230 million individuals per year in developing countries. It is caused by the trematode parasite worms, of the genus schistosoma. The disease is classified into two major types; (a) intestinal schistosomiasis caused by S. guineensis , S. mekongi , S. mansoni , S. japonicum , and S. intercalatum , and (b) urogenital schistosomiasis caused by S. haematobium . Rapid multiplication and transmission of the schistosoma parasites lead to more infected people each year. Furthermore, there is a high mortality rate of this disease, above 200,000 deaths per year in sub-Saharan Africa [ 5 ]. Shistosomiasis infection happens when parasites in larval forms produced by snails in fresh water, penetrate the skin, often of the feet, that is exposed to infested water. The larva then turns into mature schistosomes in the body. The adult female worms can produce eggs in the blood vessels causing immune reactions and damaging tissues while others get out of the body through urine or faeces.

In women, urogenital schistosomiasis can be considered as one of the risk factors for being infected by human immunodeficiency virus (HIV). The body does not react to the worm but to the worm’s eggs. Intestinal schistosomiasis can cause abdominal pain, diahrrea, enlargement of the liver, blood in the stool, fluid accumulation in the peritoneal cavity, abdominal hypertension, and spleen enlargement. Urogenital schistosomiasis, on the other hand, can cause haematuria, bladder and ureter fibrosis, damage to kidney, and bladder cancer (as a possible late-stage complication).

Microscopic identification of eggs in specimens of stool or urine is the most efficient technique for Schistosomiasis diagnosis. For detection of shistosomiamansoni and shistosomiajaponicum, only microbiological methods such as the Kato-Katz technique can be applied. The severity of infection rate is determined based on the number eggs per gram of stool sample [ 109 ]. However, this diagnostic test is time-consuming, costly, and requires the skill of a trained parasitologist; all of which limit its utility in the developing world.

5.1. Biomedical Engineering Approaches

The following sections demonstrate the most common biomedical engineering approaches for diagnosis and prognosis of schistosomiasis including: bioelectrical impedance analysis (BIA), ultrasound, computerized tomography (CT), and magnetic resonance imaging (MRI). Imaging techniques are considered as one of the most rapid, reliable, and accurate methods for detection of schistosomiasis.

5.1.1. Bioelectric Impedance Analysis (BIA)

BIA is an efficient method to assess the total body water (TBW) and its distribution between ECW (Extracellular Water) and ICW (Intracellular Water), and it has been shown that BIA is sensitive to TBW percentage alterations and the ECW to ICW ratio [ 110 ] that can provide an easy, inexpensive analysis of body composition. In De Lorenzo et al. ’s [ 110 ] study, body hydration, being defined as total body water per kg of body weight, is considerably higher in patients with schistosomiasis than in normal controls. De Lorenzo et al. [ 110 ] recruited schistosomiasis patients without clinical symptoms such as no visible fluid retention, no cardiac or renal abnormalities from the underlying disease. BIA was used to measure the amount of TBW and it was found that the percentage of body water is higher in patients with schistosomiasis than in controls (62.9% ± 3.6% vs. 57.4% ± 4.3%, p < 0.0005). Although, these patients’ anthropometric characteristic was similar with the control group, their TBW% was higher. De Lorenzo et al. [ 110 ] discusses that various reasons may have partially contributed the increased subclinical alterations in body hydration in patients with schistosomiasis. For example, the subclinical alterations may be due to the arm fat area of schistosomiasis patients being reduced to 20% lower than the control group [ 110 ]. However, it is observed that difference in TBW% is small and may not be sufficient to allow a classification.

Hypoalbuminaemia is a cause of oedema. In the absence of fluid retention, albumin levels <1.5–2 g/dL are required to have a clinically detectable increase of in ECW relative to TBW (ECW%) [ 111 ]. In De Lorenzo et al. ’s [ 110 ] findings, one of the schistosomiasis patients experienced a 1.9 g/dL and higher level with the absence of oedema. The findings suggest that albumin is vital in controlling body water homeostatis in conditions of normal and subclinically altered body hydration. The findings concluded that schistosomiasis subjects show an apparent subclinical increase in body hydration which is related to the prediction of TBW from BIA.

5.1.2. Ultrasound Imaging of Hepatosplenic-Schistosomiasis

Ultrasound (US) is known to be one of the most effective methods for investigating liver and spleen abnormalities caused by schistosomamansoni. Portal hypertension induced by the periportal fibrosis is shown to be one of the main characteristics of the hepatosplenic form of schistosomiasis, it can be demonstrated by ultrasound imaging through measurement of the diameter of the portal vein and its main tributaries (splenic vein and superior mesenteric vein). Hepatosplenic schistosomiasis can induce formation of collateral vessels in the gastrointestinal tract leading to acute bleeding or possibly death. This phenomenon can be demonstrated by ultrasound imaging indicating an increased risk of bleeding ( Figure 9 ). Ultrasound can also be applied as a diagnostic tool in the assessment of treatment efficacy and regression of fibrosis in schistosomiasis [ 112 , 113 ].

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g009.jpg

Ultrasound B-scans of the abdomen showing changes caused by schistosomiasis. White arrow: Central periportal fibrosis, red arrows: fibrosis on the periphery of the liver in a patient diagnosed with advanced hepatosplenic schistosomiasis Reproduced with permission [ 112 ].

5.1.3. Computerized Tomography (CT) and Magnetic Resonance Imaging (MRI)

Diagnosis of hepatosplenic Schistomiasis can be obtained by observing characteristic changes of the attenuation in the computerized tomography (CT) or signal intensity and enhancement pattern of the magnetic resonance images (MRI) [ 112 , 114 ]. The CT and MRI findings are usually compared and are in agreement with the ultrasound findings. In CT, low-density periportal zones are demonstrated as periportal thickening when compared with echogenic layers in ultrasound images. These low-intensity zones extend uniformly throughout the liver lobes and are enhanced after an intravenous injection of contrast agents to become either homogenous with or denser than the surrounding hepatic tissue.

In MRI, on the other hand, periportal bands are demonstrated as areas of hyperintensity on T1-weighted sequences and hyperintensity on T2-weighted sequences after the intravenous injection. Studies of acute cases applying MRI demonstrated that regions of periportal fibrosis were hyperintense on T2-weighted images in all patients ( Figure 10 ). MRI has been known as a highly responsive imaging technique for many diffused liver pathologies. Unlike ultrasound imaging, MRI is not a dynamic method and also it is not dependent on the examiner.

An external file that holds a picture, illustration, etc.
Object name is sensors-15-06947-g010.jpg

MRI: Gamma-Gandy bodies (siderotic nodules) pointed by arrows labelled “i” in the spleen of a patient diagnosed with hepatosplenicschistosomiasis. Arrow labelled “ii” shows the portal vein. Reproduced with permission [ 112 ].

In Japan, ultrasound, CT and MRI techniques and endoscopic examinations with biopsies have been used to detect lesions related to schistosmiases japonica [ 114 ]. Portal hypertension is detected by CT, Ultrsound and gastroscopic examination. The MRI imaging diagnoses were performed on chronic lesions and/or sequelae of schistosemiasis and in well-equipped hospitals because of its high cost and longer time of examination.

5.2. Summary

The summary of biomedical engineering techniques in schistosomiasis disease is listed in Table 7 .

Summary of the biomedical engineering techniques in schistosomiasis disease.

6. Lymphatic Filariasis

Lymphatic Filariasis which is also known as elephantiasis, is responsible for more than 40 million disfigured and infirm people, and intimidating more than 1.3 billion people in 72 countries [ 8 ]. The disease starts when an infected mosquito bites a human and transmits filarial parasites to the human lymphatic vessels. Then, three types of filarial worms develop inside the infected body: Wuchereria bancrofti , Brugia malayi and Brugia timori . These worms settle in the lymphatic system for 6–8 years. During this incubation period, they produce large numbers of microlariae (larvae) that interfere with the immune system. Although infection typically takes place during the childhood period, painful disfiguration will occur later in life [ 8 ].

There are two main approaches for diagnosing lymphatic filariasis: microscopic examination and serological tests. The first technique is to examine blood smears under the microscope to detect microfilariaes. The blood must be collected at night; since the microfilariaes circulate in blood only at night. Although this technique has a low sensitivity, it is still considered as an acceptable alternative diagnostic method when antigen testing cannot be done [ 115 ]. On the other hand, serological tests are based on the fact that filarial infection increases the concentrations of antifilarial IgG and IgE in the blood. Thus, one can identify antifilarial immunoglobulin type G and E antibodies (IgG, IgE) or filarial antigens using ELISA which leads to a higher sensitivity [ 115 ].

6.1. Biomedical Engineering Approaches

Few techniques have been investigated in order to overcome the limitations of lymphatic filariasis diagnosis. The following sections show the available technologies of ultrasonography, lymphoscintigraphy and immunochromatography.

6.1.1. Ultrasonography

Ultrasonography has been used to locate and study the movements of active adult filarial worms of Wuchereria bancrofti in the scrotal region in men as a diagnostic tool for lymphatic filariasis; while only a few cases of adult filariae are observed by ultrasonography in women. This movement of Wuchereria bancrofti is described as “filaria dance”; which is considered a sign of the presence of Wuchereria bancrofti worms [ 116 ].

Dreyer et al. [ 117 ] used longitudinal ultrasound examination to investigate the microfilaria dance sign in the scrotal lymphatic vessels. They found out that ultrasound is a beneficial tool to evaluate in a direct and fast way the usefulness of antifilarial drugs. In another study done by Mand et al. [ 116 ], they used a portable ultrasound unit, a 7.5 MHz linear transducer and a 3.5 MHz curved array transducer to examine different positions in subjects’ bodies; in order to locate worm nests. The findings showed that ultrasonography is suitable for diagnosis of Wuchereria bancrofti infection in female patients as well as males.

6.1.2. Lymphoscintigraphy

Radioisotope imaging (scintigraphy) was demonstrated by Freedman et al. [ 118 ] as an effective way to evaluate severity of the lymphatic filariasis; however, the primary diagnosis must be done by non-imaging laboratory assays.

6.1.3. Immunochromatography

Weil et al. [ 119 ] described an immunochromatographic filarial antigen test using specific monoclonal and polyclonal antibodies. The test is done by adding the serum onto a pink sample pad which contains dried polyclonal antifilarial antibodies attached to colloidal gold. Then two drops of a wash buffer are added to a separate wash pad. If the serum is infected, the labelled antibody and filarial antigen move to the top of the card through a nitrocellulose strip resulting in a visible pink line. This technique does not need professional operators; since its easy procedure can be done by individuals without much training. Moreover, the results are accessible within only few minutes, making this test a potentially fast diagnostic tool for lymphatic filariasis infection.

6.2. Summary

The summary of biomedical engineering techniques in lymphatic filariasis disease is listed in Table 8 .

Summary of biomedical engineering techniques in lymphatic filariasis disease.

The first recognition of Ebola virus infection occurred in 1976 in Zaire, Africa and the mortality rate was 88% [ 7 ]. There are four types of Ebola virus including: Zaire, Sudan, Ivory Coast, and Reston; the first two being responsible for the majority of hemorrhagic human fever resulting in the high mortality rate [ 120 ]. Ebola can be confused with other diseases such as malaria, since the most frequent symptoms include: fever, weakness, vomiting, diarrhea, sore throat and general aches [ 121 ]. The formal laboratory diagnostic technique is an ELISA, but this technique is not sensitive enough for all different stages of disease and is not easily accessible in countries endemic for Ebola infection [ 122 ].

7.1. Biomedical Engineering Approaches

Although, ELISA is the routine test for Ebola, two techniques, RT-PCR (reverse transcriptase polymerase chain reaction) and an optical biosensor have led to high sensitivity and specificity.

Optical Immunosensor

An optical immunosensor based on the photo immobilization technique to detect the two most common types of Ebola virus, Zaire and Sudan was developed by Petrosova et al. [ 120 ]. The sensor was fabricated by coating the tip of a fiber optic with a 200 nm layer of indium tin oxide (ITO), and later with polypyrrole benzophenone. The tip is then irradiated with a 345 nm wavelength light at intensity of 80 mW·cm −2 to excite the benzophenone radicals to bind to the Ebola virus antigen. The fiber tip then goes through a process similar to a standard ELISA. The detection is then performed via chemiluminescence measurements using a photo multiplier tube sensor. The results show that the immunosensor is able to detect titers as low as 1:960,000 and 1:1,000,000 respectively for the Zaire and Sudan strains.

7.2. Summary

The summary of the biomedical engineering technique in Ebola disease is listed in Table 9 .

Summary of the biomedical engineering technique in Ebola disease.

Leprosy is a chronic infectious disease caused by Mycobacterium leprae. The cause of the disease was identified in 1873 by G.H.A. Hansen and is also known as Hansen’s disease [ 123 ]. It remains a public health problem and transmission continues, although prevalence has been reduced in the past half century. Never the less leprosy remains as a leading infectious disease causing disabilities. At the beginning of 2011, the global registered prevalence of leprosy from 130 territories stood around 192,000 cases, while during 2010 the number of newly detected cases was about 228,000 [ 9 ]. The primary external sign of leprosy is skin lesions that if left untreated, can be progressive and cause permanent damage to the limbs, eyes, nerves and skin. Thus early diagnosis and treatment is essential [ 124 ].

Mycobacterium leprae cannot be cultured in the laboratory, hence scientists are using innovative techniques to measure the interaction of host cells with Mycobacterium leprae, however there still is no primary method for prevention of leprosy nor a means of diagnostic and prognostic testing that is practical in routine clinical care [ 125 ]. Problems with diagnosis of leprosy are related to its insidious nature with sometimes conflicting immunological, clinical and pathological manifestations even though there is minimal genetic variation among Mycobacterium leprae isolates worldwide.

An effective diagnosis method for all stages of the diseases is electroneuromyography (ENMG). ENMG is a technique where the muscle nerve under study is stimulated with electric current. The effectiveness is demonstrated by the presence of electroneuromyographic alterations among 98% of leprosy confirmed patients, especially in the diagnosis of pure neural leprosy [ 126 ].

8.1. Biomedical Engineering Approaches

Although highly sensitive and specific diagnostic tests for Mycobacterium leprae have yet to be established. Imaging techniques and biosensors, especially those based on employing DNA based markers, are considered as strong promising leprosy diagnostic techniques.

Diagnostic Imaging

The incidence of bone lesions in leprosy is low, but the radiologic findings of chronic and acute osteomyelitis that are similar to lesions of other granulomatous infectious agents are seen. Employing imaging techniques various degrees of reabsorption of the body associated with leprosy involving feet and hands with loss of disorganizing arthropathies and digits in small joints can be visualized [ 126 , 127 ].

Magnetic resonance imaging (MRI) and ultrasonography (US) have been used as imaging techniques for evaluation of pure neural leprosy. These imaging techniques employ specific markers and they have been used to assist in determining diagnosis and prognosis. US provides high resolution images of the morphological alternations in peripheral nerves; however the value of this technique in the diagnosis of peripheral neuropathy is poorly understood. Moreover, the application of MRI in leprosy peripheral nerve involvement has been described in the literature, and there have been studies of leprosy diagnosis using these techniques [ 127 ]. Goulart & Goulart [ 126 ] have shown that MRI and Doppler ultrasonography have a sensitivity of 92% and 74%, respectively, to identify active reversal reactions, based on the observation of the color flow signals. However, in case of tender neuropathy, MRI may exclude nerve abscess, while US examination can be done more rapid and effective than MRI imaging.

8.2. Summary

The summary of the biomedical engineering technique in leprosy disease is listed in Table 10 .

Summary of the biomedical engineering technique in leprosy disease.

9. Leishmaniasis

Leishmaniasis is a neglected parasitic disease that is caused by the protozoan parasite of the Leishmania genus and spread by the bite of female phlebotomine sand flies [ 12 , 128 ]. These flies inject the infective stage of promastigote parasites into human hosts [ 10 ]. These are then phagocytized by macrophages and transformed into amastigotes which will develop in the infected cells and cause skin lesions that will take months or even years to heal [ 10 ] or which if left untreated, may be fatal [ 128 ]. Leishmaniasis is a public health problem that has being reported to have a 500% increasing incidence in the most endemic areas in Africa, Asia, the Middle East, and the Mediterranean [ 10 , 11 ].

The established gold standards for Leishmaniasis detection involve the isolation of parasites either microscopically or by culture [ 12 ]. Existing available diagnostic tests include histology, culture, molecular techniques, Leishmanin skin test and serologic skin tests. Never the less those diagnostic tests have their own limitations particularly due to false positives with Chagas disease and also to a lesser extent to tuberculosis and leprosy [ 10 , 12 ].

9.1. Biomedical Engineering Approaches

To overcome the limitations of current diagnostic techniques, integrated electronic biosensors systems that are suitable for field diagnosis are being developed [ 12 ].

Impedance-Based Biosensor

Perinoto et al. [ 12 ] developed a nanostructured biosensor system to detect specific anti-leishmania antibodies by measuring the capacitance of proteoliposomes integrated onto gold interdigitated electrodes. These electrodes were functionalized by attaching antigenic proteins to the surface through repeated immersing of the electrodes into a solution containing the antigenic proteins. Electrical capacitance measurements of this structure containing immobilized proteoliposomes allow the recognition of specific anti-Leishmania amazonensis antibodies at the concentration of 10 −5 mg/mL [ 12 ]. This technique can be applied for the diagnosis of other protozoan and bacterial infectious diseases such as the tuberculosis, malaria, filariasis, schistosomiasis, onchocerciasis and other neglected diseases.

9.2. Summary

The summary of the biomedical engineering technique in Leishmaniasis disease is listed in Table 11 .

Summary of the biomedical engineering technique in Leishmaniasis disease.

10. American Trypanosomiasis (Chagas Disease)

American trypanosomiasis, or Chagas disease is endemic in tropical areas and is caused by the parasite Trypanosome cruzi. The parasite is transmitted by the triatomine family of bugs (commonly known as “kissing bugs”) that act as reservoirs and vectors in endemic areas. As of 2002, 5–6 million people were infected worldwide, with 25 million more at risk [ 129 ]. The Disease is typically found in Latin America, however, due to population mobility, the disease has been increasingly detected in North America, Europe and Western Pacific countries [ 129 ].

There are two phases of the disease: acute and chronic. The acute phase starts when the parasite Trypanosome cruzi enters the body, and may last from 4 to 8 weeks. Local reaction at the point of entry such as skin irritation occurs due to the bug bite or entry through the conjunctiva. Once Trypanosome cruzi enters the body, the patient may experience common symptoms such as fever, vomiting, diarrhoea, and anorexia. During this phase, a patient experiences acute myocarditis, where an ECG may show signs of sinus tachycardia, first degree atrio-ventricular (A-V) block, low QRS voltage, or primary T-wave changes. The chronic phase starts when the parasite can no longer be detected in the blood, and symptoms of myocarditis and meningoencephalitis disappear. In this phase, detection can only be done through serological tests such as IgG antibody detection. In the chronic phase, 10%–30% of patients may suffer cardiac damage, while others may experience damage to the cardiac conduction network and the autonomic nervous system [ 129 , 130 ].

For diagnosis of the acute phase of infection, parasites can be detected by direct parasitological tests. Thin and thick blood smears allow microscopic observation of the parasite. [ 129 , 130 ]. In the chronic phase, only 50% of the patients may show positive results using parasitological tests. Serological tests such as haemagglutination, indirect immuno fluorescence, and ELISA are more applicable. Current approaches of detection have limitations in terms of consistency of results, and requiring at least two techniques to be positive in order to properly diagnose Chagas infection. However, as untreated Chagas may lead to damage to the cardiac conduction and autonomic nervous system, biomedical engineering approaches have been investigated and developed for a more accurate diagnosis of the disease.

10.1. Biomedical Engineering Approaches

In order to overcome the difficulties faced in Chagas diagnosis, biomedical engineers have developed more accurate tools to help diagnose and identify the disease. This instrumentation consists of biosensors that are based on electric impedance and current measurements.

A Chagas disease biosensor has been reported by Diniz et al. [ 131 ]. It had gold and platinum electrodes that were coated with an oxide layer and recombinant antigens (cytoplasmic repetitive antigen and flagellar repetitive antigen for Chagas), and dipped in patient’s serum. The study showed that complex impedance increases much faster when the electrodes are dipped in Chagas positive serum, compared to a slower increase in impedance when dipped in Chagas negative serum.

In a similar study by Ribone et al. [ 132 ], a biosensor utilizing gold electrodes was constructed. Similar to the Chagas ELISA method, the electrode is covered with Trypanosome cruzi antigen that reacts with patients’ anti-Trypanosome cruzi IgG. A classic enzyme label using horseradish peroxidase conjugated with anti-human IgG was applied. However, instead of using the typical optical detection method, a potential of less than 0.1 V was applied and electrode current was measured. An increase in current indicates an increase in IgG antibody. In a separate study by Belluzo et al. [ 133 ], the same biosensor was improvised with custom made chimeric receptors that react with the IgG antibody. The biosensor was shown to achieve 100% specificity while attaining increased sensitivity with a detection limit that is eight times lower compared to commercially available ELISA kits.

10.2. Summary

The summary of biomedical engineering techniques in Chagas disease is listed in Table 12 .

Summary of biomedical engineering techniques in Chagas disease.

11. Conclusions and Outlook

This paper has reviewed the biomedical engineering approaches for diagnosing and monitoring tropical diseases. Methods such as ultrasound, echocardiography and electrocardiography, strain gauge plethysmography, laser Doppler and bioelectrical impedance and other techniques allow a noninvasive approach to tropical disease diagnosis and management while overcoming the limitations of conventional approaches. Clinical decision support systems based on self-organizing maps, multilayer feed-forward neural networks and adaptive neuro-fuzzy inference systems assist clinicians to obtain a more specific diagnosis of the disease type and stage, whereas lab-on-chip and micro/nanofluidics and photonic approaches allow for a more sensitive and specific diagnosis at lower cost and a shorter turnaround time. Although there has been much progress, there are still many opportunities to improve these current biomedical engineering approaches as well as to develop new approaches for the described and other tropical diseases.

Acknowledgments

This research is supported by University of Malaya High Impact Research Grant UM-MOHE UM.C/625/1/HIR/MOHE05 from Ministry of Higher Education Malaysia (MOHE), Fundamental Research Grant Scheme (FRGS: FP042-2013B) and University of Malaya Research Grant (UMRG: RG009A-13AET). Fatimah Ibrahim would like to acknowledge Yayasan Sultan Iskandar Johore for supporting and funding the one off special equipment in dengue research.

Conflicts of Interest

The authors declare no conflict of interest.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • View all journals
  • Explore content
  • About the journal
  • Publish with us
  • Sign up for alerts
  • Review Article
  • Published: 10 October 2018

Artificial intelligence in healthcare

  • Kun-Hsing Yu   ORCID: orcid.org/0000-0001-9892-8218 1 ,
  • Andrew L. Beam 1 &
  • Isaac S. Kohane 1 , 2  

Nature Biomedical Engineering volume  2 ,  pages 719–731 ( 2018 ) Cite this article

41k Accesses

1247 Citations

99 Altmetric

Metrics details

  • Computational models
  • Machine learning
  • Predictive medicine

Artificial intelligence (AI) is gradually changing medical practice. With recent progress in digitized data acquisition, machine learning and computing infrastructure, AI applications are expanding into areas that were previously thought to be only the province of human experts. In this Review Article, we outline recent breakthroughs in AI technologies and their biomedical applications, identify the challenges for further progress in medical AI systems, and summarize the economic, legal and social implications of AI in healthcare.

This is a preview of subscription content, access via your institution

Access options

Access Nature and 54 other Nature Portfolio journals

Get Nature+, our best-value online-access subscription

24,99 € / 30 days

cancel any time

Subscribe to this journal

Receive 12 digital issues and online access to articles

92,52 € per year

only 7,71 € per issue

Rent or buy this article

Prices vary by article type

Prices may be subject to local taxes which are calculated during checkout

recent research papers of biomedical engineering

Simonite, T. Google’s AI eye doctor gets ready to go to work in India. WIRED (6 August 2017).

Lee, R., Wong, T. Y. & Sabanayagam, C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis. 2 , 17 (2015).

Article   Google Scholar  

Lin, D. Y., Blumenkranz, M. S., Brothers, R. J. & Grosvenor, D. M. The sensitivity and specificity of single-field nonmydriatic monochromatic digital fundus photography with remote image interpretation for diabetic retinopathy screening: a comparison with ophthalmoscopy and standardized mydriatic color photography. Am. J. Ophthalmol. 134 , 204–213 (2002).

Article   PubMed   Google Scholar  

Zheng, Y., He, M. & Congdon, N. The worldwide epidemic of diabetic retinopathy. Indian J. Ophthalmol. 60 , 428–431 (2012).

Article   PubMed   PubMed Central   Google Scholar  

Gulshan, V. et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 316 , 2402–2410 (2016).

Poplin, R. et al. Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning. Nat. Biomed. Eng. 2 , 158–164 (2018).

Abràmoff, M. D., Lavin, P. T., Birch, M., Shah, N. & Folk, J. C. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digit. Med. 1 , 39 (2018).

Russell, S. J. & Norvig, P. Artificial Intelligence: A Modern Approach (Prentice Hall, New Jersey, 2010).

Krizhevsky, A., Sutskever, I. & Hinton, G. E. in Advances in Neural Information Processing Systems 1097–1105 (Curran Associates, Nevada, 2012).

Lewis-Kraus, G. The great A.I. awakening. The New York Times Magazine (14 December 2016).

Kundu, M., Nasipuri, M. & Basu, D. K. Knowledge-based ECG interpretation: a critical review. Pattern Recognit. 33 , 351–373 (2000).

Jha, S. & Topol, E. J. Adapting to artificial intelligence: radiologists and pathologists as information specialists. JAMA 316 , 2353–2354 (2016).

Golub, T. R. et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286 , 531–537 (1999).

Article   CAS   PubMed   Google Scholar  

Wang, Y. et al. Gene selection from microarray data for cancer classification—a machine learning approach. Comput. Biol. Chem. 29 , 37–46 (2005).

Article   PubMed   CAS   Google Scholar  

Yu, K. H. et al. Predicting ovarian cancer patients’ clinical response to platinum-based chemotherapy by their tumor proteomic signatures. J. Proteome Res. 15 , 2455–2465 (2016).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Yu, K. H. et al. Omics AnalySIs System for PRecision Oncology (OASISPRO): a web-based omics analysis tool for clinical phenotype prediction. Bioinformatics 34 , 319–320 (2017).

Article   CAS   PubMed Central   Google Scholar  

Check Hayden, E. The automated lab. Nature 516 , 131–132 (2014).

Miller, R. A. Medical diagnostic decision support systems–past, present, and future: a threaded bibliography and brief commentary. J. Am. Med. Inform. Assoc. 1 , 8–27 (1994).

Musen, M. A., Middleton, B. & Greenes, R. A. in Biomedical Informatics (eds Shortliffe, E. H. & Cimino, J. J.) 643–674 (Springer, London, 2014).

Shortliffe, E. Computer-Based Medical Consultations: MYCIN Vol. 2 (Elsevier, New York, 2012).

Chapter   Google Scholar  

Szolovits, P., Patil, R. S. & Schwartz, W. B. Artificial intelligence in medical diagnosis. Ann. Intern. Med. 108 , 80–87 (1988).

de Dombal, F. T., Leaper, D. J., Staniland, J. R., McCann, A. P. & Horrocks, J. C. Computer-aided diagnosis of acute abdominal pain. Br. Med. J. 2 , 9–13 (1972).

Shortliffe, E. H. et al. Computer-based consultations in clinical therapeutics: explanation and rule acquisition capabilities of the MYCIN system. Comput. Biomed. Res. 8 , 303–320 (1975).

Barnett, G. O., Cimino, J. J., Hupp, J. A. & Hoffer, E. P. DXplain. An evolving diagnostic decision-support system . JAMA 258 , 67–74 (1987).

Miller, R. A., McNeil, M. A., Challinor, S. M., Masarie, F. E. Jr & Myers, J. D. The INTERNIST-1/QUICK MEDICAL REFERENCE Project — status report. Western J. Med. 145 , 816–822 (1986).

Berner, E. S. et al. Performance of four computer-based diagnostic systems. N. Engl. J. Med. 330 , 1792–1796 (1994).

Szolovits, P. & Pauker, S. G. Categorical and probabilistic reasoning in medical diagnosis. Artif. Intell. 11 , 115–144 (1978).

Deo, R. C. Machine learning in medicine. Circulation 132 , 1920–1930 (2015).

Yu, K. H. & Snyder, M. Omics profiling in precision oncology. Mol. Cell. Proteomics 15 , 2525–2536 (2016).

Roberts, K. et al. Biomedical informatics advancing the national health agenda: the AMIA 2015 year-in-review in clinical and consumer informatics. J. Am. Med. Inform. Assoc. 24 , 185–190 (2017).

Cloud AutoML ALPHA (Google Cloud); https://cloud.google.com/automl/

Goodfellow, I., Bengio, Y., Courville, A. & Bengio, Y. Deep Learning 1 (MIT Press, Cambridge, 2016).

Google Scholar  

Gill, N. S. Overview and applications of artificial neural networks. Xenonstack https://www.xenonstack.com/blog/data-science/artificial-neural-networks-applications-algorithms/ (2017).

TOP500 List – November 2006 (TOP500); https://www.top500.org/list/2006/11/

Beam, A. L. & Kohane, I. S. Translating artificial intelligence into clinical care. JAMA 316 , 2368–2369 (2016).

Kamentsky, L. et al. Improved structure, function and compatibility for CellProfiler: modular high-throughput image analysis software. Bioinformatics 27 , 1179–1180 (2011).

Ching, T. et al. Opportunities and obstacles for deep learning in biology and medicine. J. R. Soc. Interface 15 , 20170387 (2018).

Tomczak, K., Czerwinska, P. & Wiznerowicz, M. The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge. Contemp. Oncol. 19 , 68–77 (2015).

Sudlow, C. et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 12 , e1001779 (2015).

Ljosa, V., Sokolnicki, K. L. & Carpenter, A. E. Annotated high-throughput microscopy image sets for validation. Nat. Methods 9 , 637 (2012).

Williams, E. et al. The image data resource: a bioimage data integration and publication platform. Nat. Methods 14 , 775–781 (2017).

DesRoches, C. M. et al. Electronic health records in ambulatory care–a national survey of physicians. N. Engl. J. Med. 359 , 50–60 (2008).

Hsiao, C. J. et al. Office-based physicians are responding to incentives and assistance by adopting and using electronic health records. Health Aff. 32 , 1470–1477 (2013).

Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 542 , 115–118 (2017).

Beck, A. H. et al. Systematic analysis of breast cancer morphology uncovers stromal features associated with survival. Sci. Transl. Med. 3 , 108ra113 (2011).

Yu, K. H. et al. Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. Nat. Commun. 7 , 12474 (2016).

Shademan, A. et al. Supervised autonomous robotic soft tissue surgery. Sci. Transl. Med. 8 , 337ra364 (2016).

Reed, J. C. Chest Radiology: Plain Film Patterns and Differential Diagnoses (Elsevier Health Sciences, Philadelphia, 2010).

Lodwick, G. S., Haun, C. L., Smith, W. E., Keller, R. F. & Robertson, E. D. Computer diagnosis of primary bone tumors: a preliminary report. Radiology 80 , 273–275 (1963).

van Ginneken, B., Setio, A. A., Jacobs, C. & Ciompi, F. Off-the-shelf convolutional neural network features for pulmonary nodule detection in computed tomography scans. In IEEE 12th International Symposium Biomedical Imaging (ISBI) 286–289 (IEEE, 2015).

Lakhani, P. & Sundaram, B. Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology 284 , 574–582 (2017).

Wang, X. et al. ChestX-ray8: hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. Preprint at https://arxiv.org/abs/1705.02315 (2017).

Yao, L. et al. Learning to diagnose from scratch by exploiting dependencies among labels. Preprint at https://arxiv.org/abs/1710.10501 (2017).

Rajpurkar, P. et al. CheXNet: radiologist-level pneumonia detection on chest X-rays with deep learning. Preprint at https://arxiv.org/abs/1711.05225 (2017).

Samala, R. K. et al. Mass detection in digital breast tomosynthesis: deep convolutional neural network with transfer learning from mammography. Med. Phys. 43 , 6654–6666 (2016).

Arevalo, J., González, F. A., Ramos-Pollán, R., Oliveira, J. L. & Lopez, M. A. G. Convolutional neural networks for mammography mass lesion classification. In IEEE 37th Annual International Conference of the Engineering in Medicine and Biology Society (EMBC) 797–800 (IEEE, 2015).

510(k) Premarket Notification (FDA, 2017); https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K163253

Marr, B. First FDA approval for clinical cloud-based deep learning in healthcare. Forbes (20 January 2017).

Rigel, D. S., Friedman, R. J., Kopf, A. W. & Polsky, D. ABCDE—an evolving concept in the early detection of melanoma. Arch. Dermatol. 141 , 1032–1034 (2005).

Thomas, L. et al. Semiological value of ABCDE criteria in the diagnosis of cutaneous pigmented tumors. Dermatology 197 , 11–17 (1998).

Ercal, F., Chawla, A., Stoecker, W. V., Lee, H. C. & Moss, R. H. Neural network diagnosis of malignant melanoma from color images. IEEE Trans. Biomed. Eng. 41 , 837–845 (1994).

Wolf, J. A. et al. Diagnostic inaccuracy of smartphone applications for melanoma detection. JAMA Dermatol. 149 , 422–426 (2013).

Panwar, N. et al. Fundus photography in the 21st century — a review of recent technological advances and their implications for worldwide healthcare. Telemed. J. E. Health 22 , 198–208 (2016).

American Diabetes Association. 10. Microvascular complications and foot care. Diabetes Care 40 , 88–98 (2017).

Menke, A., Casagrande, S., Geiss, L. & Cowie, C. C. Prevalence of and trends in diabetes among adults in the United States, 1988–2012. JAMA 314 , 1021–1029 (2015).

Abràmoff, M. D. et al. Improved automated detection of diabetic retinopathy on a publicly available dataset through integration of deep learning. Investigative Opthalmology Visual Sci. 57 , 5200–5206 (2016).

Rorke, L. B. Pathologic diagnosis as the gold standard. Cancer 79 , 665–667 (1997).

Lakhani, S. R. & Ashworth, A. Microarray and histopathological analysis of tumours: the future and the past? Nat. Rev. Cancer 1 , 151–157 (2001).

Rubegni, P. et al. Automated diagnosis of pigmented skin lesions. Int. J. Cancer 101 , 576–580 (2002).

Stang, A. et al. Diagnostic agreement in the histopathological evaluation of lung cancer tissue in a population-based case-control study. Lung Cancer 52 , 29–36 (2006).

Yu, K. H. et al. Association of omics features with histopathology patterns in lung adenocarcinoma. Cell Syst. 5 , 620–627 (2017).

Litjens, G. et al. Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis. Sci. Rep. 6 , 26286 (2016).

Bejnordi, B. E. et al. Machine learning detection of breast cancer lymph node metastases. JAMA 318 , 2199–2210 (2017).

Cireşan, D. C., Giusti, A., Gambardella, L. M. & Schmidhuber, J. in Medical Image Computing and Computer-Assisted Intervention — MICCAI 2013 (eds Mori, K. et al.) 411–418 (Springer, Berlin, Heidelberg, 2013).

Manak, M. S. et al. Live-cell phenotypic-biomarker microfluidic assay for the risk stratification of cancer patients via machine learning. Nat. Biomed. Eng. https://doi.org/10.1038/s41551-018-0285-z (2018).

Robboy, S. J. et al. Pathologist workforce in the United States: I. Development of a predictive model to examine factors influencing supply. Arch. Pathol. Lab. Med. 137 , 1723–1732 (2013).

Litjens, G. et al. A survey on deep learning in medical image analysis. Med. Image Anal. 42 , 60–88 (2017).

Quang, D., Chen, Y. & Xie, X. DANN: a deep learning approach for annotating the pathogenicity of genetic variants. Bioinformatics 31 , 761–763 (2015).

Quang, D. & Xie, X. DanQ: a hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences. Nucleic Acids Res. 44 , e107 (2016).

Article   PubMed   PubMed Central   CAS   Google Scholar  

DePristo, M. & Poplin, R. DeepVariant: highly accurate genomes with deep neural networks. Google AI Blog https://research.googleblog.com/2017/12/deepvariant-highly-accurate-genomes.html (2017).

Poplin, R. et al. Creating a universal SNP and small indel variant caller with deep neural networks. Preprint at https://www.biorxiv.org/content/early/2016/12/14/092890 (2018).

Kamps, R. et al. Next-generation sequencing in oncology: genetic diagnosis, risk prediction and cancer classification. Int. J. Mol. Sci. 18 , 308 (2017).

Article   PubMed Central   CAS   Google Scholar  

He, Z. & Yu, W. Stable feature selection for biomarker discovery. Comput. Biol. Chem. 34 , 215–225 (2010).

Zhang, Z. et al. Three biomarkers identified from serum proteomic analysis for the detection of early stage ovarian cancer. Cancer Res. 64 , 5882–5890 (2004).

Wallden, B. et al. Development and verification of the PAM50-based Prosigna breast cancer gene signature assay. BMC Med. Genomics 8 , 54 (2015).

Sweeney, T. E., Wong, H. R. & Khatri, P. Robust classification of bacterial and viral infections via integrated host gene expression diagnostics. Sci. Transl. Med. 8 , 346ra391 (2016).

Article   CAS   Google Scholar  

Huang, T., Hoffman, B., Meschino, W., Kingdom, J. & Okun, N. Prediction of adverse pregnancy outcomes by combinations of first and second trimester biochemistry markers used in the routine prenatal screening of Down syndrome. Prenat. Diagn. 30 , 471–477 (2010).

PubMed   Google Scholar  

Mook, S. et al. Metastatic potential of T1 breast cancer can be predicted by the 70-gene MammaPrint signature. Ann. Surg. Oncol. 17 , 1406–1413 (2010).

Farina, D. et al. Man/machine interface based on the discharge timings of spinal motor neurons after targeted muscle reinnervation. Nat. Biomed. Eng. 1 , 0025 (2017).

Altman, R. B. Artificial intelligence (AI) systems for interpreting complex medical datasets. Clin. Pharmacol. Ther. 101 , 585–586 (2017).

Cai, X. et al. Real-time prediction of mortality, readmission, and length of stay using electronic health record data. J. Am. Med. Inform. Assoc. 23 , 553–561 (2016).

Makar, M., Ghassemi, M., Cutler, D. M. & Obermeyer, Z. Short-term mortality prediction for elderly patients using medicare claims data. Int. J. Mach. Learn. Comput. 5 , 192–197 (2015).

Ng, T., Chew, L. & Yap, C. W. A clinical decision support tool to predict survival in cancer patients beyond 120 days after palliative chemotherapy. J. Palliat. Med. 15 , 863–869 (2012).

Delen, D., Oztekin, A. & Kong, Z. J. A machine learning-based approach to prognostic analysis of thoracic transplantations. Artif. Intell. Med. 49 , 33–42 (2010).

Churpek, M. M. et al. Predicting cardiac arrest on the wards: a nested case-control study. Chest 141 , 1170–1176 (2012).

Churpek, M. M. et al. Multicenter development and validation of a risk stratification tool for ward patients. Am. J. Respir. Crit. Care Med. 190 , 649–655 (2014).

Lundberg, S. M. et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat. Biomed. Eng. https://doi.org/10.1038/s41551-018-0304-0 (2018).

Li, X. et al. Digital health: tracking physiomes and activity using wearable biosensors reveals useful health-related information. PLoS Biol. 15 , e2001402 (2017).

Majumder, S., Mondal, T. & Deen, M. J. Wearable sensors for remote health monitoring. Sensors 17 , 130 (2017).

Article   PubMed Central   Google Scholar  

Pastorino, M., Arredondo, M., Cancela, J. & Guillen, S. Wearable sensor network for health monitoring: the case of Parkinson disease. J. Phys. Conf. Ser. 450 , 012055 (2013).

Mercer, K., Li, M., Giangregorio, L., Burns, C. & Grindrod, K. Behavior change techniques present in wearable activity trackers: a critical analysis. JMIR Mhealth Uhealth 4 , e40 (2016).

Takacs, J. et al. Validation of the Fitbit One activity monitor device during treadmill walking. J. Sci. Med. Sport 17 , 496–500 (2014).

Yang, R., Shin, E., Newman, M. W. & Ackerman, M. S. When fitness trackers don’t ‘fit’: end-user difficulties in the assessment of personal tracking device accuracy. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing 623–634 (ACM, 2015).

Endeavour Partners. Inside wearables: how the science of human behavior change offers the secret to long-term engagement. Medium https://blog.endeavour.partners/inside-wearable-how-the-science-of-human-behavior-change-offers-the-secret-to-long-term-engagement-a15b3c7d4cf3 (2017).

Herz, J. C. Wearables are totally failing the people who need them most. Wired (11 June 2014).

Clawson, J., Pater, J. A., Miller, A. D., Mynatt, E. D. & Mamykina, L. No longer wearing: investigating the abandonment of personal health-tracking technologies on Craigslist. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing 647–658 (ACM, 2015).

Wheeler, M. J. Overview on robotics in the laboratory. Ann. Clin. Biochem. 44 , 209–218 (2007).

Moustris, G. P., Hiridis, S. C., Deliparaschos, K. M. & Konstantinidis, K. M. Evolution of autonomous and semi-autonomous robotic surgical systems: a review of the literature. Int. J. Med. Robot . 7 , 375–392 (2011).

Gomes, P. Surgical robotics: reviewing the past, analysing the present, imagining the future. Robot. Comput. Integr. Manuf. 27 , 261–266 (2011).

Majdani, O. et al. A robot-guided minimally invasive approach for cochlear implant surgery: preliminary results of a temporal bone study. Int. J. Comput. Assist. Radiol. Surg. 4 , 475–486 (2009).

Elek, R. et al. Recent trends in automating robotic surgery. In 2016 IEEE 20th Jubilee International Conference on Intelligent Engineering Systems (INES) 27–32 (IEEE, 2016).

Liew, C. The future of radiology augmented with artificial intelligence: a strategy for success. Eur. J. Radiol. 102 , 152–156 (2018).

Jones, L., Golan, D., Hanna, S. & Ramachandran, M. Artificial intelligence, machine learning and the evolution of healthcare: a bright future or cause for concern? Bone Joint Res. 7 , 223–225 (2018).

Obermeyer, Z. & Emanuel, E. J. Predicting the future — big data, machine learning, and clinical medicine. N. Engl. J. Med. 375 , 1216–1219 (2016).

Krause, J. et al. Grader variability and the importance of reference standards for evaluating machine learning models for diabetic retinopathy. Ophthalmology 125 , 1264–1272 (2018).

Rebholz-Schuhmann, D. et al. The CALBC silver standard corpus for biomedical named entities—a study in harmonizing the contributions from four independent named entity taggers. In LREC 568–573 (2010).

Kirby, J. C. et al. PheKB: a catalog and workflow for creating electronic phenotype algorithms for transportability. J. Am. Med. Inform. Assoc. 23 , 1046–1052 (2016).

Simonyan, K., Vedaldi, A. & Zisserman, A. Deep inside convolutional networks: visualising image classification models and saliency maps. Preprint at https://arxiv.org/abs/1312.6034 (2013).

Ribeiro, M. T., Singh, S. & Guestrin, C. “Why should I trust you?”: explaining the predictions of any classifier. Preprint at https://arxiv.org/abs/1602.04938 (2016).

LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521 , 436–444 (2015).

Boulanger-Lewandowski, N., Bengio, Y. & Vincent, P. Modeling temporal dependencies in high-dimensional sequences: application to polyphonic music generation and transcription. Preprint at https://arxiv.org/abs/1206.6392 (2012).

Zoph, B. & Le, Q. V. Neural architecture search with reinforcement learning. Preprint at https://arxiv.org/abs/1611.01578 (2016).

Lee, L. M. & Gostin, L. O. Ethical collection, storage, and use of public health data: a proposal for a national privacy protection. JAMA 302 , 82–84 (2009).

Narayan, S., Gagné, M. & Safavi-Naini, R. Privacy preserving EHR system using attribute-based infrastructure. In Proceedings of the 2010 ACM Workshop on Cloud Computing Security Workshop 47–52 (ACM, 2010).

Dolin, R. H. et al. HL7 Clinical Document Architecture, Release 2. J. Am. Med. Inform. Assoc. 13 , 30–39 (2006).

Mandl, K. D. & Kohane, I. S. Escaping the EHR trap—the future of health IT. N. Engl. J. Med. 366 , 2240–2242 (2012).

Mandel, J. C., Kreda, D. A., Mandl, K. D., Kohane, I. S. & Ramoni, R. B. SMART on FHIR: a standards-based, interoperable apps platform for electronic health records. J. Am. Med. Inform. Assoc. 23 , 899–908 (2016).

All eyes are on AI. Nat. Biomed. Eng . 2 , 139 (2018).

Yu, K. H. & Kohane I. S. Framing the challenges of artificial intelligence in medicine. BMJ Qual. Safety https://doi.org/10.1136/bmjqs-2018-008551 (2018).

Dignum, V. Ethics in artificial intelligence: introduction to the special issue. Ethics Inf. Technol . 20 , 1–3 (2018).

Price, I. & Nicholson, W. Artificial Intelligence in Health Care: Applications and Legal Implications (Univ. Michigan Law School, 2017).

Mukherjee, S. A.I. versus M.D. What happens when diagnosis is automated? The New Yorker (3 April 2017).

Del Beccaro, M. A., Jeffries, H. E., Eisenberg, M. A. & Harry, E. D. Computerized provider order entry implementation: no association with increased mortality rates in an intensive care unit. Pediatrics 118 , 290–295 (2006).

Longhurst, C. A. et al. Decrease in hospital-wide mortality rate after implementation of a commercially sold computerized physician order entry system. Pediatrics 126 , 14–21 (2010).

Carspecken, C. W., Sharek, P. J., Longhurst, C. & Pageler, N. M. A clinical case of electronic health record drug alert fatigue: consequences for patient outcome. Pediatrics 131 , 1970–1973 (2013).

Ash, J. S., Berg, M. & Coiera, E. Some unintended consequences of information technology in health care: the nature of patient care information system-related errors. J. Am. Med. Inform. Assoc. 11 , 104–112 (2004).

Lehman, C. D. et al. Diagnostic accuracy of digital screening mammography with and without computer-aided detection. JAMA Intern. Med. 175 , 1828–1837 (2015).

Koppel, R. et al. Role of computerized physician order entry systems in facilitating medication errors. JAMA 293 , 1197–1203 (2005).

Middleton, B. et al. Enhancing patient safety and quality of care by improving the usability of electronic health record systems: recommendations from AMIA. J. Am. Med. Inform. Assoc. 20 , 2–8 (2013).

Gottlieb, S. Twitter (12 April 2018); https://twitter.com/SGottliebFDA/status/984378648781312002

Digital Health Software Precertification (Pre-Cert) Program (FDA); https://www.fda.gov/MedicalDevices/DigitalHealth/DigitalHealthPreCertProgram/default.htm

Estrin, D. & Sim, I. Open mHealth architecture: an engine for health care innovation. Science 330 , 759–760 (2010).

Shortliffe, E. H. Computer programs to support clinical decision making. JAMA 258 , 61–66 (1987).

Armbruster, D. A., Overcash, D. R. & Reyes, J. Clinical chemistry laboratory automation in the 21st century—amat victoria curam (victory loves careful preparation). Clin. Biochem. Rev. 35 , 143–153 (2014).

Rosenfeld, L. A golden age of clinical chemistry: 1948–1960. Clin. Chem. 46 , 1705–1714 (2000).

Kuperman, G. J. et al. Medication-related clinical decision support in computerized provider order entry systems: a review. J. Am. Med. Inform. Assoc. 14 , 29–40 (2007).

Glassman, P. A., Simon, B., Belperio, P. & Lanto, A. Improving recognition of drug interactions: benefits and barriers to using automated drug alerts. Med. Care 40 , 1161–1171 (2002).

FDA permits marketing of artificial intelligence algorithm for aiding providers in detecting wrist fractures. https://www.fda.gov/newsevents/newsroom/pressannouncements/ucm608833.htm (FDA, 2018).

Haenssle, H. A. et al. Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. Ann. Oncol. 29 , 1836–1842 (2018).

Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. Preprint at https://arxiv.org/abs/1409.1556 (2014).

Murphy, K. P. & Bach F. Machine Learning: A Probabilistic Perspective (MIT Press, Cambridge, 2012).

Download references

Acknowledgements

K.-H.Y. is supported by a Harvard Data Science Postdoctoral Fellowship. I.S.K. was supported in part by the NIH grant OT3OD025466. Figure 4 was generated by using the computational infrastructure supported by the AWS Cloud Credits for Research, the Microsoft Azure Research Award, and the NVIDIA GPU Grant Programme.

Author information

Authors and affiliations.

Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA

Kun-Hsing Yu, Andrew L. Beam & Isaac S. Kohane

Boston Children’s Hospital, Boston, MA, USA

Isaac S. Kohane

You can also search for this author in PubMed   Google Scholar

Contributions

K.-H.Y. conceived and designed the article, performed the literature review and wrote and revised the manuscript. A.L.B. and I.S.K. edited the manuscript. I.S.K. supervised the work.

Corresponding author

Correspondence to Isaac S. Kohane .

Ethics declarations

Competing interests.

Harvard Medical School (K.-H.Y.) submitted a provisional patent application on digital pathology profiling to the United States Patent and Trademark Office (USPTO).

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Cite this article.

Yu, KH., Beam, A.L. & Kohane, I.S. Artificial intelligence in healthcare. Nat Biomed Eng 2 , 719–731 (2018). https://doi.org/10.1038/s41551-018-0305-z

Download citation

Received : 12 December 2017

Accepted : 05 September 2018

Published : 10 October 2018

Issue Date : October 2018

DOI : https://doi.org/10.1038/s41551-018-0305-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Medical image fusion based on machine learning for health diagnosis and monitoring of colorectal cancer.

  • Yifeng Peng
  • Haijun Deng

BMC Medical Imaging (2024)

SyntDiaNet: Integrating feature extraction, transfer learning and classifier-embedded generative adversarial network for advanced pneumonia diagnosis

  • Rahul Gowtham Poola
  • Siva Sankar Yellampalli

Multimedia Tools and Applications (2024)

Ethical concerns around privacy and data security in AI health monitoring for Parkinson’s disease: insights from patients, family members, and healthcare professionals

  • Martin J. McKeown

AI & SOCIETY (2024)

AI-driven surveillance of the health and disease status of ocean organisms: a review

  • Arghya Mandal
  • Apurba Ratan Ghosh

Aquaculture International (2024)

A deep learning based fine-grained classification algorithm for grading of visual impairment in cataract patients

  • Jiewei Jiang
  • Zhongwen Li

Optoelectronics Letters (2024)

Quick links

  • Explore articles by subject
  • Guide to authors
  • Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

recent research papers of biomedical engineering

Book cover

  • Conference proceedings
  • © 2021

Advances in Biomedical Engineering and Technology

Select Proceedings of ICBEST 2018

  • Albert A. Rizvanov 0 ,
  • Bikesh Kumar Singh 1 ,
  • Padma Ganasala 2

Kazan Federal University, Kazan, Russia

You can also search for this editor in PubMed   Google Scholar

Department of Biomedical Engineering, National Institute of Technology Raipur, Raipur, India

Gayatri vidya parishad college of engineering, visakhapatnam, india.

Presents latest research findings in biomedical engineering and technology

Covers a wide range of topics relevant to academics and medical practitioners

Includes contributions from global experts in bio-engineering

Part of the book series: Lecture Notes in Bioengineering (LNBE)

Conference series link(s): ICBEST: International Conference on Biomedical Engineering Science and Technology

25k Accesses

81 Citations

Conference proceedings info: ICBEST 2018.

  • Table of contents

About this book

Editors and affiliations, about the editors, bibliographic information.

  • Publish with us

Buying options

  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
  • Durable hardcover edition

Tax calculation will be finalised at checkout

Other ways to access

This is a preview of subscription content, log in via an institution to check for access.

Table of contents (39 papers)

Front matter, extraction and phytochemical analysis of coccinia indica fruit using uv-vis and ftir spectroscopy.

  • Alok Sharma, Bidyut Mazumdar, Amit Keshav

Numerical Simulation Method to Predict Air Flow and Contaminant Control in a Multiple Bed Intensive Care Unit of Hospital

  • Arvind Kumar Sahu, Shobha Lata Sinha, Tikendra Nath Verma

A Simple Robust Image Processing Algorithm for Analysis of Static Foot Pressure Intensity Image to Detect Foot Risk Areas in Diabetic Patients

  • Hari S. Nair, Navya Thomas, R. Periyasamy

Heavy Metal Ions Detection by Carbon Paste Electrode as an Electrochemical Sensor

  • Arti Mourya, Bidyut Mazumdar, Sudip K. Sinha

Median Filtering Detection Using Markov Process in Digital Images

  • Saurabh Agarwal, Satish Chand

Differential of EMG Activity of Selected Calf Muscle During DLHR Exercise in Relation to Performance Level

  • Monika, L. M. Saini, Saravjeet Singh

Advanced Encryption Standard Algorithm in Multimodal Biometric Image

  • Sharmila S. More, Bhawna Narain, B. T. Jadhav

In-Silico Construction of Hybrid ORF Protein to Enhance Algal Oil Content for Biofuel

  • Mohit Nigam, Ruchi Yadav, Garima Awasthi

ANFIS Detects the Changes in Stressful Patterns of Sleep EEG

  • Prabhat Kumar Upadhyay, Chetna Nagpal

Recent Advances in Deep Learning Techniques and Its Applications: An Overview

  • Abhishek Hazra, Prakash Choudhary, M. Sheetal Singh

Validation of a New Method of Pediatric Refraction: Large Aperture Lens Rack

  • Anupam Sahu, Samrat Chatterjee, Deepshikha Agrawal, Pradeep Chand Dubey

Comparative Evaluation of in Vitro Antioxidant, Amylase Inhibition and Cytotoxic Activity of Cur-Pip Dual Drug Loaded Nanoparticles

  • Trilochan Satapathy, Prasanna Kumar Panda, Gitanjali Mishra

Improved ERP Classification Algorithm for Brain–Computer Interface of ALS Patient

  • Vyom Raj, Shreya Sharma, Mridu Sahu, Samrudhi Mohdiwale

Size Reduction in Multiband Planar Antenna for Wireless Applications Using Current Distribution Technique

  • Pravin Tajane, P. L. Zade

Classification of Hepatic Disease Using Machine Learning Algorithms

  • Lokesh Singh, Rekh Ram Janghel, Satya Prakash Sahu

Anti-hyperlipidemic and Antioxidant Activities of a Combination of Terminalia Arjuna and Commiphora Mukul on Experimental Animals

  • Jhakeshwar Prasad, Ashish Kumar Netam, Trilochan Satapathy, S. Prakash Rao, Parag Jain

Epileptic Seizure Detection Using Deep Recurrent Neural Networks in EEG Signals

  • Archana Verma, Rekh Ram Janghel

Detection of Disease from Leaf of Vegetables and Fruits Using Deep Learning Technique

  • Avisha Jaiswal, Saurabh Pathak, Yogesh Kumar Rathore, Rekh Ram Janghel

Evaluation for Toxicity and Improved Therapeutic Effectiveness of Natural Polymer Co-administered Along with Venocin in Acetic Acid-Induced Colitis Using Rat Model

  • Ashish Kumar Netam, Jhakeshwar Prasad, Trilochan Satapathy, Parag Jain
  • Biomechanics
  • Biomaterials
  • Tissue Engineering
  • Biomedical Signal Processing
  • Biomedical Instrumentation and Medical Devices
  • Telemedicine and Internet of Things
  • Biomedical Image Processing
  • ICBEST 2018

Albert A. Rizvanov

Bikesh Kumar Singh

Padma Ganasala

Book Title : Advances in Biomedical Engineering and Technology

Book Subtitle : Select Proceedings of ICBEST 2018

Editors : Albert A. Rizvanov, Bikesh Kumar Singh, Padma Ganasala

Series Title : Lecture Notes in Bioengineering

DOI : https://doi.org/10.1007/978-981-15-6329-4

Publisher : Springer Singapore

eBook Packages : Biomedical and Life Sciences , Biomedical and Life Sciences (R0)

Copyright Information : Springer Nature Singapore Pte Ltd. 2021

Hardcover ISBN : 978-981-15-6328-7 Published: 29 September 2020

Softcover ISBN : 978-981-15-6331-7 Published: 30 September 2021

eBook ISBN : 978-981-15-6329-4 Published: 28 September 2020

Series ISSN : 2195-271X

Series E-ISSN : 2195-2728

Edition Number : 1

Number of Pages : XXII, 488

Number of Illustrations : 48 b/w illustrations, 157 illustrations in colour

Topics : Regenerative Medicine/Tissue Engineering , Biomedical Engineering and Bioengineering , Biomaterials , Signal, Image and Speech Processing , Imaging / Radiology

Policies and ethics

  • Find a journal
  • Track your research

IEEE EMBS

Comprehensive Study from 50 Top Scientists Details Five Cutting-edge Advances in Biomedical Engineering and Their Applications in Medicine

  • February 21, 2024

Alliance of 50 experts from 34 elite universities reveals pioneering engineering advance across five vital domains.

PISCATAWAY, N.J.– (BUSINESS WIRE) – IEEE, the world’s largest technical professional organization dedicated to advancing technology for humanity, and the IEEE Engineering in Medicine and Biology Society (IEEE EMBS), today published a detailed position paper on the field of biomedical engineering titled, “Grand Challenges at the Interface of Engineering and Medicine.” The paper, published in the IEEE Open Journal of Engineering in Medicine and Biology (IEEE OJEMB), was written by a consortium of 50 renowned researchers from 34 prestigious universities around the world, and lays the foundation for a concerted worldwide effort to achieve technological and medical breakthroughs.

“What we’ve accomplished here will serve as a roadmap for groundbreaking research to transform the landscape of medicine in the coming decade,” said Dr. Michael Miller, senior author of the paper and professor and director of the Department of Biomedical Engineering at Johns Hopkins University. “The outcomes of the task force, featuring significant research and training opportunities, are poised to resonate in engineering and medicine for decades to come.”

“This paper represents a major milestone in the advancement of biomedical engineering, which could only have been achieved through close collaboration rather than the work of many siloed individuals,” said consortium member Dr. Metin Akay, founding chair of the Biomedical Engineering Department at the University of Houston and Ambassador of IEEE EMBS. “We have a shared commitment to advancing patient-centric technologies, and healthcare efficacy and accessibility — which extends beyond academic institutions — and elevating healthcare quality, reducing costs and improving lives worldwide.”

The position paper was the result of a two-day workshop organized by IEEE EMBS and the Department of Biomedical Engineering at Johns Hopkins University and the Department of Bioengineering at the University of California San Diego. Through the course of the workshop, the researchers identified five primary medical challenges that have yet to be addressed, but, by solving them with advanced biomedical engineering approaches, can greatly improve human health. By focusing on these five areas, the consortium has laid out a roadmap for future research and funding.

The Five Grand Challenges Facing Biomedical Engineering 1. Bridging precision engineering and precision medicine for personalized physiology avatars

In an increasingly digital age, we have technologies that gather immense amounts of data on patients, which clinicians can add to or pull from. Making use of this data to develop accurate models of physiology, called “avatars” — which take into account multimodal measurements and comorbidities, concomitant medications, potential risks and costs — can bridge individual patient data to hyper-personalized care, diagnosis, risk prediction, and treatment. Advanced technologies, such as wearable sensors and digital twins, can provide the basis of a solution to this challenge.

2. The pursuit of on-demand tissue and organ engineering for human health

Tissue engineering is entering a pivotal period in which developing tissues and organs on demand, either as permanent or temporary implants, is becoming a reality. To shepherd the growth of this modality, key advancements in stem cell engineering and manufacturing — along with ancillary technologies such as gene editing — are required. Other forms of stem cell tools, such as organ-on-a-chip technology, can soon be built using a patient’s own cells and can make personalized predictions and serve as “avatars.”

3. Revolutionizing neuroscience using artificial intelligence (AI) to engineer advanced brain-interface systems

Using AI, we have the opportunity to analyze the various states of the brain through everyday situations and real-world functioning to noninvasively pinpoint pathological brain function. Creating technology that does this is a monumental task, but one that is increasingly possible. Brain prosthetics, which supplement, replace or augment functions, can relieve the disease burden caused neurological conditions. Additionally, AI modeling of brain anatomy, physiology, and behavior, along with the synthesis of neural organoids, can unravel the complexities of the brain and bring us closer to understanding and treating these diseases.

4. Engineering the immune system for health and wellness

With a heightened understanding of the fundamental science governing the immune system, we can strategically make use of the immune system to redesign human cells as therapeutic and medically invaluable technologies. The application of immunotherapy in cancer treatment provides evidence of the integration of engineering principles with innovations in vaccines, genome, epigenome and protein engineering, along with advancements in nanomedicine technology, functional genomics and synthetic transcriptional control.

5. Designing and engineering genomes for organism repurposing and genomic perturbations

Despite the rapid advances in genomics in the past few decades, there are obstacles remaining in our ability to engineer genomic DNA. Understanding the design principles of the human genome and its activity can help us create solutions to many different diseases that involve engineering new functionality into human cells, effectively leveraging the epigenome and transcriptome, and building new cell-based therapeutics. Beyond that, there are still major hurdles in gene delivery methods for in vivo gene engineering, in which we see biomedical engineering being a component to the solution to this problem.

“These grand challenges offer unique opportunities that can transform the practice of engineering and medicine,” remarked Dr. Shankar Subramaniam, lead author of the taskforce, distinguished professor of Bioengineering at the University of California San Diego and past President of IEEE EMBS. “Innovations in the form of multi-scale sensors and devices, creation of humanoid avatars and the development of exceptionally realistic predictive models driven by AI can radically change our lifestyles and response to pathologies. Institutions can revolutionize education in biomedical and engineering, training the greatest minds to engage in the most important problem of all times — human health.”

This position paper was published in IEEE OJEMB, and can be accessed at this link .

“We have a shared commitment to advancing patient-centric technologies, and healthcare efficacy and accessibility, especially for major healthcare challenges such as chronic conditions, substance abuse and mental disorders.”

About the IEEE Open Journal of Engineering in Medicine and Biology

The IEEE Open Journal of Engineering in Medicine and Biology covers developing and applying engineering concepts and methods to biology, medicine, and health sciences to effectively solve biological, medical, and healthcare problems. For more information about OJEMB, please visit www.embs.org/ojemb .

About the IEEE Engineering in Medicine and Biology Society

The IEEE Engineering in Medicine and Biology Society (EMBS) is the world’s largest international society of biomedical engineers. With more than 12,000 members residing in some 97 countries around the globe, IEEE EMBS is a truly global connection, fostering fellowship and providing access to fascinating people, best practices, new information, innovative ideas, and a variety of expert opinions from one of science’s fastest growing fields: biomedical engineering. From formalized mathematical theory through experimental science and from technological development to practical clinical applications, IEEE EMBS members support scientific, technical, and educational activities as they apply to the concepts and methods of the physical and engineering sciences in biology and medicine. By working together, we can transform and revolutionize the future of medicine and healthcare. For more information about the IEEE EMBS, please visit www.embs.org .

COMMENTS

  1. Biomedical engineering

    Biomedical engineering is a branch of engineering that applies principles and design concepts of engineering to healthcare. Biomedical engineers deal with medical devices such as imaging...

  2. Nature Biomedical Engineering

    Review Article 20 Apr 2023 Current issue Hail the AI journal editor Editorial 22 Jan 2024 Designing stable humanized antibodies Alissa M. Hummer Charlotte M. Deane News & Views 27 Dec 2023...

  3. biomedical engineering Latest Research Papers

    biomedical engineering Latest Research Papers | ScienceGate biomedical engineering Recently Published Documents TOTAL DOCUMENTS 3863 (FIVE YEARS 921) H-INDEX 43 (FIVE YEARS 11) Latest Documents Most Cited Documents Contributed Authors Related Sources Related Keywords

  4. IEEE Transactions on Biomedical Engineering

    IEEE Transactions on Biomedical Engineering. null | IEEE Xplore. Need Help? US & Canada: +1 800 678 4333 Worldwide: +1 732 981 0060 Contact & Support

  5. Modern Trends and Applications of Intelligent Methods in Biomedical

    The Special Issue "Modern Trends and Applications of Intelligent Methods in Biomedical Signal and Image Processing" is aimed at the new proposals and intelligent solutions that constitute the state of the art of the intelligent methods for biomedical data processing from selected areas of signal and image processing.

  6. Home

    Journal updates Research on Biomedical Engineering is an online, peer-reviewed journal dedicated to all fields of Biomedical Engineering. Editor-in-Chief Alcimar Barbosa Soares Submission to first decision (median) The Brazilian Society of Biomedical Engineering (opens in a new tab) View all volumes and issues Submission guidelines

  7. Bioengineering for global health

    Bioengineering for global health. Bioengineering has the power to improve health globally by engineering diagnostic, treatment and disease monitoring platforms that function in diverse settings ...

  8. Smart bioelectronics and biomedical devices

    The recent research focuses on sweat sensors with a diverse form factors and materials. The review also covers different mechanisms of sensing as well as their advantages and challenges. Lim et al. [ 39 ] discussed the requirement of bioreactor design for tissue engineering, which should provide a biomimetic physiological platform by optimizing ...

  9. Home page

    BioMedical-Engineering OnLine has published the following thematic series: International Conference on Aging, Innovation and Rehabilitation Advanced signal processing and modeling for neuroengineering Trends in e-health and m-health: technology developments for improved healthcare Artificial intelligence in biomedical imaging

  10. Recent findings and applications of biomedical engineering for COVID-19

    Recent findings and applications of biomedical engineering for COVID-19 diagnosis: a critical review Le Minh Bui , Huong Thi Thu Phung , Thuy-Tien Ho Thi , Vijai Singh , Rupesh Maurya , Khushal Khambhati , Pages 8594-8613 | Received 10 Aug 2021, Accepted 28 Sep 2021, Published online: 18 Oct 2021 Cite this article

  11. Recent Advances in Biopolymeric Composite Materials for Tissue

    Recent Advances in Biopolymeric Composite Materials for Tissue Engineering and Regenerative Medicines: A Review Muhammad Umar Aslam Khan, 1,2,3,* Saiful Izwan Abd Razak, 2,4 Wafa Shamsan Al Arjan, 5 Samina Nazir, 5 T. Joseph Sahaya Anand, 6 Hassan Mehboob, 7 and Rashid Amin 8,* Samy Madbouly, Academic Editor

  12. Biomedical Engineering in Human Health

    This Section of Biomedicines aims to rapidly publish reviews and original peer-reviewed papers describing research on the development and application principles of biomedical engineering in human health and diseases. The main topics of the Section include but are not limited to: Tissue engineering; Neural engineering; Genetic engineering;

  13. Biomedical engineering and ethics: reflections on medical devices and

    The unprecedented high number of retracted papers can be a proxy for the high number of low-quality research on COVID-19. For this reason, a rapid search for papers regarding COVID-19 or SARS-CoV-2, and the previous epidemic/pandemics (i.e., SARS, MERS, Swine Flu) as a comparison, was performed both on OvidSP and the Retraction Watch Database.

  14. Advances and challenges in Bioinformatics and Biomedical Engineering

    This Supplement issue, presents five research articles which are distributed, mainly due to the subject they address, from the 8th International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO 2020), which was held on line, during September, 30th-2nd October, 2020. These contributions have been chosen because of their quality and the importance of their findings.

  15. An engineering makeover of biomedical research

    An engineering makeover of biomedical research. The clinical translation of research findings into improvements in human health can take decades. Engineers perform biomedical research with a ...

  16. Bioengineering

    Feature papers represent the most advanced research with significant potential for high impact in the field. ... articles related to the advancement and development of novel mathematical models broadly applied across a wide range of biomedical engineering and medical physics, including diagnostic, therapeutic, imaging, and interventional ...

  17. Biomedical Engineering News

    Silk Solutions: Unlocking the Secrets of Natural Materials To Weave a Better World. Professor Benedetto Marelli develops silk-based technologies with uses "from lab to fork," including helping crops grow and preserving perishable foods. Growing up in Milan, Benedetto…. Technology November 11, 2023.

  18. Five Cutting-edge Advances in Biomedical Engineering and Their

    Revolutionizing neuroscience by using AI to engineer advanced brain interface systems. Engineering the immune system for health and wellness. Designing and engineering genomes for organism repurposing and genomic perturbations. These are the five research areas where the field of biomedical engineering has the potential to achieve tremendous ...

  19. A Review-Biomedical Engineering-Present and Future Prospective

    Prominent biomedical engineering applications include the development of biocompatible prostheses, various diagnostic and therapeutic medical devices ranging from clinical equipment to...

  20. Research Areas

    Hopkins BME faculty and students focus their research efforts on seven main areas. Learn more about each focus area and the work that is being conducted. ... Johns Hopkins Biomedical Engineering. Research. Research Areas. ... Design, build, and test new imaging technologies to improve disease diagnosis and guide clinical procedures. Learn More.

  21. The Application of Biomedical Engineering Techniques to the Diagnosis

    Recent research and studies in the past decade have proposed and introduced various biomedical engineering approaches that attempt to address the issues faced in the diagnosis of tropical diseases. This review paper focuses on current biomedical engineering approaches for the more prominent diseases of dengue, malaria, cholera, schistosomiasis ...

  22. Artificial intelligence in healthcare

    Nature Biomedical Engineering - This Review summarizes the medical applications of artificial intelligence, and its economic, legal and social implications for healthcare. ... recent AI research ...

  23. Advances in Biomedical Engineering and Technology

    This book comprises select peer-reviewed papers presented at the International Conference on Biomedical Engineering Science and Technology: Roadway from Laboratory to Market (ICBEST 2018) organized by Department of Biomedical Engineering, National Institute of Technology Raipur, Chhattisgarh, India. The book covers latest research in a wide ...

  24. IEEE Reviews in Biomedical Engineering

    Need Help? US & Canada: +1 800 678 4333 Worldwide: +1 732 981 0060 Contact & Support

  25. Hopkins BME takes on grand challenges in biomedical engineering

    IEEE, the world's largest technical professional organization dedicated to advancing technology for humanity, and the IEEE Engineering in Medicine and Biology Society (IEEE EMBS), today published a detailed position paper on the field of biomedical engineering titled, "Grand Challenges at the Interface of Engineering and Medicine." The paper, published in the IEEE Open Journal of ...

  26. Comprehensive Study from 50 Top Scientists Details Five Cutting-edge

    PISCATAWAY, N.J.- (BUSINESS WIRE) - IEEE, the world's largest technical professional organization dedicated to advancing technology for humanity, and the IEEE Engineering in Medicine and Biology Society (IEEE EMBS), today published a detailed position paper on the field of biomedical engineering titled, "Grand Challenges at the Interface of Engineering and Medicine."

  27. Biomedical Engineering Society on Instagram: "Submit your paper to the

    16 likes, 0 comments - bmesociety on May 6, 2023: "Submit your paper to the new Special Issue on Graduate Education in Biomedical Engineering! Pap..." Biomedical Engineering Society on Instagram: "Submit your paper to the new Special Issue on Graduate Education in Biomedical Engineering!