Published on in Vol 25 (2023)

Preprints (earlier versions) of this paper are available at, first published .
Digital Education for the Deployment of Artificial Intelligence in Health Care

Digital Education for the Deployment of Artificial Intelligence in Health Care

Digital Education for the Deployment of Artificial Intelligence in Health Care


1Ophthalmology Department, Sao Paulo Federal University, Sao Paulo, Brazil

2Laboratory for Computational Physiology, Massachusetts Institute of Technology, Cambridge, MA, United States

3University of the Philippines College of Medicine, Manila, Philippines

4Escuela Superior de Medicina, Instituto Politecnico Nacional, Mexico City, Mexico

5Department of Biostatistics, Harvard TH Chan School of Public Health, Boston, MA, United States

Corresponding Author:

Luis Filipe Nakayama, MD

Laboratory for Computational Physiology

Massachusetts Institute of Technology

77 Massachusetts Ave

Cambridge, MA, 02139

United States

Phone: 1 617 253 7818


Artificial Intelligence (AI) represents a significant milestone in health care's digital transformation. However, traditional health care education and training often lack digital competencies. To promote safe and effective AI implementation, health care professionals must acquire basic knowledge of machine learning and neural networks, critical evaluation of data sets, integration within clinical workflows, bias control, and human-machine interaction in clinical settings. Additionally, they should understand the legal and ethical aspects of digital health care and the impact of AI adoption. Misconceptions and fears about AI systems could jeopardize its real-life implementation. However, there are multiple barriers to promoting electronic health literacy, including time constraints, overburdened curricula, and the shortage of capacitated professionals. To overcome these challenges, partnerships among developers, professional societies, and academia are essential. Integrating specialists from different backgrounds, including data specialists, lawyers, and social scientists, can significantly contribute to combating digital illiteracy and promoting safe AI implementation in health care.

J Med Internet Res 2023;25:e43333



The health care ecosystem comprises multiple stakeholders, including, but not limited to, health care personnel (HCP), hospital managers, public and private health systems, and end users. With the emergence of artificial intelligence (AI), there is great potential to improve health care outcomes, including reduction of costs and increased access [1]. AI is a transformative technology that can improve medical decision-making, clinical diagnosis, and treatment [2,3]. Image-based diagnosis in radiology, ophthalmology, pathology, and dermatology, genome interpretation, clinical predictions, biomarker discovery, and robot surgery are examples of many AI applications in health care [3-6]. However, traditional education and training of HCP seldom encompass digital competencies [7]. As a result, professionals in the health sector who will be affected by the deployment of AI have minimal exposure to relevant digital education [8] and are currently unable to harness the full potential of implementing AI in health care. In this viewpoint article, we argue that teaching HCP about digital health care is critical for realizing the benefits of AI in health care and for the safe deployment of this technology.

Recent studies have emphasized the importance of teaching digital competencies for HCP, such as their roles in digitalization of health care, knowledge of basic computer science concepts, and legal and ethical aspects [7]. To effectively use AI, HCP will need to understand, interpret, and meaningfully critique the outputs of AI models [9]. This task demands inputs from people with diverse backgrounds such as computer science, mathematics, statistics, law, ethics, social science, and health care. This includes traditional health competencies such as clinical skills and epidemiology [10].

To address these needs, the proposed new medical specialty of “Clinical AI” would expand the more traditional specialty of clinical informatics [11]. Clinical AI specialists would have a leading role in the decentralized approach to safer AI deployment and regulation; they would also continuously review and recalibrate AI models [12]. It has become evident that digital health care teaching is not prevalent in most health-related schools’ curricula [7]; however, incorporating digital health care teaching into the existing curricula presents significant challenges, including an already overburdened curriculum, compartmentalization of the educational program, and time constraints [8]. Knowledge of digital health care is also scarce among fully trained HCP due to cultural unreadiness and a gap between early and late adopters, among other reasons [8].

From our standpoint, electronic health literacy is fundamental for not only the workforce but also other stakeholders in the health care ecosystem, including the end user. Misconceptions and unfounded fears from HCP and patients may jeopardize the real-life implementation of AI systems in health care. Building trust and refuting false beliefs are essential for successful deployment. Moreover, social scientists must analyze the sociocultural implications of software, wearables, and self-care technologies, which are crucial to understanding and avoiding biases and dangerous AI results, ensuring safe implementation [13].

HCP must adapt to the changes brought about by the integration of AI in health care. Incorporating digital health competencies in HCP training curricula is undoubtedly challenging, and some approaches have been proposed, including classes, web-based courses, and certifications [8]. Trainees should be taught fundamental AI concepts including taxonomy [9], general aspects of data sets, integration within clinical workflows, concepts on biases, the value of clinical deployment, human-machine interaction in clinical settings, and specific health care applications of AI [8,9,12,14,15]. Werner et al [16] recently reported the successful implementation of a longitudinal, modular course on digital health in medical graduation with positive student feedback. The course included a modularly structured core curriculum and elective courses, beginning with principles of scientific methods, an orientation phase, and concluding with each student selecting a specialization area and preparing a research project as an independent academic achievement [16].

HCP must learn how to collaborate with professionals from diverse backgrounds and how to engage partners outside the system, such as the developers of electronic medical records [9]. They should also learn to access and generate open access data sets for secondary data analysis—such a step being fundamental to the reduction of AI biases and the promotion of fair and generalizable models [17,18].

Since medical training curricula are already saturated, practical training in “Applied AI” has been proposed as a feasible approach to saving time [19]. It has also been proposed that medical information that was once memorized but is now available through AI algorithms should be less emphasized in favor of digital health skills that enable safe and effective interaction with AI technologies [19]. Critical analysis of AI studies, including the ability to identify relevant research questions and recognize the quality of applied data sets, is fundamental [15]. It is also essential to understand the system’s inputs and outputs, metrics, external validation [15], the adequacy of the chosen operating threshold [20], intended use, and epidemiological and socioeconomic considerations. Finally, practical aspects of implementing electronic health literacy into current systems, such as postdeployment studies and recalibration, must be considered [17,21].

Digital competencies could also be taught to graduated HCP through continued education forums on digital health care, scientific meetings, conferences, and datathon and hackathon events. These activities enable collaborative exchanges between HCP and data specialists [22].

Patients play a critical role in the successful deployment of health care AI. Despite the reported benefit of AI adoption in multiple fields, patients' misconceptions and false beliefs can lead to mistrust in such systems [5]. For example, a recent survey found that most patients reported being uncomfortable receiving an AI-assisted diagnosis with 90% accuracy, but were incapable of explaining its rationale [23]. Therefore, it is essential to familiarize patients with the benefits and limitations of AI in health care to gain their trust and support.

To promote this goal, a combined effort involving AI developers, HCP, and patient associations could provide a venue for appraising patients on health care AI [1]. Patients not only serve as the end users of health care AI but also constitute partners in the AI enterprise. By generating greater awareness of AI, patients tend to become convinced that by sharing their data, they will improve health care for themselves and other patients [24].

As real-world implementation of AI in health care becomes widespread, greater exposure to AI-driven medical technology will increase patients' awareness and encourage them to consent to sharing their data [24]. Assertive statements from developers reassuring their commitment to data privacy are also essential to develop trust [24-26].

Electronic health literacy is crucial for harnessing the digital health care revolution. However, the promotion of digital health care is faced with several challenges. For instance, time constraint is a significant obstacle for training HCP on digital health as curricula are already saturated, and adding new content is unsustainable. One possible alternative is to integrate new competencies into existing program components [9]. Furthermore, there is a shortage of professionals with robust data and analytical skills [9], and low incentive to engage busy professionals on this topic. Educating patients on the benefits of AI tools is also challenging. Therefore, stakeholders must be informed about the potential gains, and common myths need to be deconstructed. Partnerships among developers, professional HCP societies, academia, and specialists from different backgrounds, including social scientists, can significantly contribute to advancing the agenda of combating HCP digital illiteracy and preparing the society as a whole to realize the benefits of AI implementation [8].


LFN is a researcher supported by Lemann Foundation, Instituto da Visão-IPEPO. All authors declared that they had insufficient or no funding to support open access publication of this manuscript, including from affiliated organizations or institutions, funding agencies, or other organizations. JMIR Publications provided article processing fee (APF) support for the publication of this article.

Conflicts of Interest

None declared.

  1. Matheny ME, Whicher D, Thadaney Israni S. Artificial intelligence in health care: a report from the National Academy of Medicine. JAMA. Feb 11, 2020;323(6):509-510. [CrossRef] [Medline]
  2. Meskó B, Görög M. A short guide for medical professionals in the era of artificial intelligence. NPJ Digit Med. Sep 24, 2020;3(1):126. [FREE Full text] [CrossRef] [Medline]
  3. Yu K, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nat Biomed Eng. Oct 10, 2018;2(10):719-731. [CrossRef] [Medline]
  4. Korn Malerbi F, Barreto Melo G. Feasibility of screening for diabetic retinopathy using artificial intelligence, Brazil. Bull World Health Org. Oct 01, 2022;100(10):643-647. [CrossRef]
  5. Keel S, Lee PY, Scheetz J, Li Z, Kotowicz MA, MacIsaac RJ, et al. Feasibility and patient acceptability of a novel artificial intelligence-based screening model for diabetic retinopathy at endocrinology outpatient services: a pilot study. Sci Rep. Mar 12, 2018;8(1):4330. [FREE Full text] [CrossRef] [Medline]
  6. Ruamviboonsuk P, Tiwari R, Sayres R, Nganthavee V, Hemarat K, Kongprayoon A, et al. Real-time diabetic retinopathy screening by deep learning in a multisite national screening programme: a prospective interventional cohort study. Lancet Digit Health. Apr 2022;4(4):e235-e244. [CrossRef]
  7. Aulenkamp J, Mikuteit M, Löffler T, Schmidt J. Overview of digital health teaching courses in medical education in Germany in 2020. GMS J Med Educ. 2021;38(4):Doc80. [FREE Full text] [CrossRef] [Medline]
  8. Gray K, Slavotinek J, Dimaguila GL, Choo D. Artificial intelligence education for the health workforce: expert survey of approaches and needs. JMIR Med Educ. Apr 04, 2022;8(2):e35223. [FREE Full text] [CrossRef] [Medline]
  9. Rajaram A, Moore K, Mamdani M. Preparing family medicine trainees for the information revolution: pearls, potential, promise, and pitfalls. Can Fam Physician. Jun 2019;65(6):390-392. [FREE Full text] [Medline]
  10. Paige SR, Stellefson M, Krieger JL, Anderson-Lewis C, Cheong J, Stopka C. Proposing a transactional model of eHealth literacy: concept analysis. J Med Internet Res. Oct 02, 2018;20(10):e10175. [FREE Full text] [CrossRef] [Medline]
  11. Gardner RM, Overhage JM, Steen EB, Munger BS, Holmes JH, Williamson JJ, et al. Core content for the subspecialty of clinical informatics. J Am Med Inform Assoc. Mar 01, 2009;16(2):153-157. [CrossRef]
  12. Panch T, Duralde E, Mattie H, Kotecha G, Celi LA, Wright M, et al. A distributed approach to the regulation of clinical AI. PLOS Digit Health. May 26, 2022;1(5):e0000040. [FREE Full text] [CrossRef] [Medline]
  13. Lupton D. Towards critical digital health studies: reflections on two decades of research in health and the way forward. Health (London). Jan 20, 2016;20(1):49-61. [CrossRef] [Medline]
  14. Academy of Medical Royal Colleges. URL: [accessed 2022-12-30]
  15. Ting DS, Lee AY, Wong TY. An ophthalmologist's guide to deciphering studies in artificial intelligence. Ophthalmology. Nov 2019;126(11):1475-1479. [FREE Full text] [CrossRef] [Medline]
  16. Werner R, Henningsen M, Schmitz R, Guse A, Augustin M, Gauer T. Digital Health meets Hamburg integrated medical degree program iMED: concept and introduction of the new interdisciplinary 2 track Digital Health. GMS J Med Educ. 2020;37(6):Doc61. [FREE Full text] [CrossRef] [Medline]
  17. Mitchell WG, Dee EC, Celi LA. Generalisability through local validation: overcoming barriers due to data disparity in healthcare. BMC Ophthalmol. May 21, 2021;21(1):228. [FREE Full text] [CrossRef] [Medline]
  18. Seastedt KP, Schwab P, O'Brien Z, Wakida E, Herrera K, Marcelo PGF, et al. Global healthcare fairness: we should be sharing more, not less, data. PLOS Digit Health. Oct 6, 2022;1(10):e0000102. [FREE Full text] [CrossRef] [Medline]
  19. Banerjee M, Chiew D, Patel KT, Johns I, Chappell D, Linton N, et al. The impact of artificial intelligence on clinical education: perceptions of postgraduate trainee doctors in London (UK) and recommendations for trainers. BMC Med Educ. Aug 14, 2021;21(1):429. [FREE Full text] [CrossRef] [Medline]
  20. Korn Malerbi F, Barreto Melo G. Feasibility of screening for diabetic retinopathy using artificial intelligence, Brazil. Bull World Health Org. Oct 01, 2022;100(10):643-647. [CrossRef]
  21. Thompson H, Sharma B, Bhalla S, Boley R, McCluskey C, Dligach D, et al. Bias and fairness assessment of a natural language processing opioid misuse classifier: detection and mitigation of electronic health record data disadvantages across racial subgroups. J Am Med Inform Assoc. Oct 12, 2021;28(11):2393-2403. [FREE Full text] [CrossRef] [Medline]
  22. Celi LA, Lokhandwala S, Montgomery R, Moses C, Naumann T, Pollard T, et al. Datathons and software to promote reproducible research. J Med Internet Res. Aug 24, 2016;18(8):e230. [FREE Full text] [CrossRef] [Medline]
  23. Khullar D, Casalino LP, Qian Y, Lu Y, Krumholz HM, Aneja S. Perspectives of patients about artificial intelligence in health care. JAMA Netw Open. May 02, 2022;5(5):e2210309. [FREE Full text] [CrossRef] [Medline]
  24. Upton R. Artificial Intelligence’s Need for Health Data – Finding An Ethical Balance. HIT Consultant Media. 2019. URL: [accessed 2023-06-09]
  25. Long CR, Purvis RS, Flood-Grady E, Kimminau KS, Rhyne RL, Burge MR, et al. Health researchers' experiences, perceptions and barriers related to sharing study results with participants. Health Res Policy Syst. Mar 04, 2019;17(1):25. [FREE Full text] [CrossRef] [Medline]
  26. Weitzman ER, Kaci L, Mandl KD. Sharing medical data for health research: the early personal health record experience. J Med Internet Res. May 25, 2010;12(2):e14. [FREE Full text] [CrossRef] [Medline]

AI: artificial intelligence
HCP: health care personnel

Edited by A Mavragani; submitted 08.10.22; peer-reviewed by M Murero, S Kendale; comments to author 28.12.22; revised version received 19.01.23; accepted 05.04.23; published 22.06.23.


©Fernando Korn Malerbi, Luis Filipe Nakayama, Robyn Gayle Dychiao, Lucas Zago Ribeiro, Cleva Villanueva, Leo Anthony Celi, Caio Vinicius Regatieri. Originally published in the Journal of Medical Internet Research (, 22.06.2023.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on, as well as this copyright and license information must be included.