Skip to main content
NCBI home page
Singapore Medical Journal logoLink to Singapore Medical Journal
. 2024 Mar 26;65(3):159–166. doi: 10.4103/singaporemedj.SMJ-2023-203

Medical education and physician training in the era of artificial intelligence

Yueyuan Xu 1,*, Zehua Jiang 1,2,*, Daniel Shu Wei Ting 3,4,5, Alfred Wei Chieh Kow 6, Fernando Bello 7, Josip Car 8, Yih-Chung Tham 3,4,9,†,✉, Tien Yin Wong 1,2,3,†
PMCID: PMC11060639  PMID: 38527300

Abstract

With the rise of generative artificial intelligence (AI) and AI-powered chatbots, the landscape of medicine and healthcare is on the brink of significant transformation. This perspective delves into the prospective influence of AI on medical education, residency training and the continuing education of attending physicians or consultants. We begin by highlighting the constraints of the current education model, challenges in limited faculty, uniformity amidst burgeoning medical knowledge and the limitations in ‘traditional’ linear knowledge acquisition. We introduce ‘AI-assisted’ and ‘AI-integrated’ paradigms for medical education and physician training, targeting a more universal, accessible, high-quality and interconnected educational journey. We differentiate between essential knowledge for all physicians, specialised insights for clinician-scientists and mastery-level proficiency for clinician-computer scientists. With the transformative potential of AI in healthcare and service delivery, it is poised to reshape the pedagogy of medical education and residency training.

Keywords: AI assistance and integration, medical curricula, medical education, physician training

INTRODUCTION

How should we teach our medical students and train them to be a physician? For over a century, the foundation of medical education has been grounded on the principles set forth in the Flexner report.[1] This report advocated for elevating the standards of medical students, increasing involvement of medical faculty in frontier research and forging close collaboration with hospitals to promote hands-on clinical training.

In the past few decades, in response to ageing populations globally, there is a shift of burden from acute to chronic diseases, resulting in increasing healthcare costs. Recognising these challenges, the Lancet Commission proposed a transition from a largely apprenticeship, time-based medical training programme to a more competency-based curriculum.[2] This proposed reform underscores the importance of implementing new concepts in team-based care and patient-centred approaches and integrating new technology (e.g., artificial intelligence [AI], simulations) into the curriculum.

Despite recent efforts undertaken by various medical schools and hospital systems worldwide to improve medical education and training, tangible transformations remain elusive.[3,4,5] Current medical education and residency training programmes, often spanning nearly 15 years or more, are characterised by their lengthiness, inefficiency and a linear progression from foundational knowledge to clinical experience. Such traditional models will struggle to meet the burgeoning demand for skilled physicians, especially in populous developing nations like China and India.[6,7]

The current approaches are inadequate to address the current and looming challenges of global healthcare.[8] The World Health Organization projects a staggering shortfall of 12.9 million healthcare workers by 2035.[9] Taken together, this calls for substantial transformation in our medical education and physician training frameworks.

CONSTRAINTS AND CHALLENGES

A typical model of medical education and post-medical school physician training is shown in Figure 1a. The process can be divided into two major phases: one that occurs in universities and medical schools and one that occurs in hospitals and healthcare environments. University education begins with general education, which is considered the cornerstone of the college curriculum. After 1–2 years, students gain foundational understanding of natural sciences and proceed to attend specialised courses in medical schools or enter Doctor of Medicine programmes. Here, they delve into the incremental layers of medical knowledge, spanning subjects like physiology, pathology and anatomy. Upon mastering the pathological intricacies of diseases, students then progress to clinical modules, encompassing fields like internal medicine, surgery, gynaecology and paediatrics.[10,11]

Figure 1.

Figure 1

Open in a new tab

Graphs show (a) traditional model vs. future (b) AI-assisted model and (c) AI-integrated model of medical education and physician training. AI: artificial intelligence

The traditional education model can be fragmented, prolonged and inefficient. First, the current teacher-centred approach follows a ‘one-size-fits-all’ teaching model, where teaching is conducted uniformly, without considering the unique needs and backgrounds of individual students. The quality of education may vary depending on medical school resources, faculty expertise and learning time. Second, classical textbooks often serve as the main reference teaching material. Yet, they can be outdated and misaligned with contemporary clinical practices, potentially leaving students underequipped for the dynamic landscape of disease management. Third, the resources allocated to developing a medical programme, including course materials, technology, faculty time and providing medical students with clinical exposure, can be costly. The shortage of medical educators, combined with the high costs and limited accessibility, can further exacerbate the global inconsistency in medical education quality.[12]

Upon achieving accreditation and mastery in theoretical knowledge, medical students then proceed to 5–8 years of standardised residency training. At this stage, they are fully immersed in hospital settings, where they acquire specialised knowledge and clinical expertise under supervision.[13]

Given the sheer volume of patients, residents are often overwhelmed with monotonous clinical tasks, leaving them with limited time for academic and research pursuits. Similarly, senior physicians are burdened by clinical duties and have limited time to mentor and train residents properly. Moreover, some hospitals lack standardisation in their training systems of clinical skills. The teaching proficiency of clinical faculties can vary significantly, posing challenges in ensuring high-quality residency training.[14] In addition, due to the concerns about patient safety and the skill levels of residents, they are often assigned simpler tasks, which can impede their clinical skill development.[15] Conversely, residents are expected to produce scientific publications, further straining their schedules. Altogether, the accumulated burden of intense learning and training may potentially undermine the nurturing of empathy and clinical proficiency.[16]

As residents transition into attending physicians, they gradually accumulate experience and knowledge through time-based learning and skills gained in clinical practice, eventually becoming senior doctors. However, this training process relies heavily on experience and can sometimes limit training opportunities. Occasionally, it might even compel junior doctors to undertake responsibilities that surpass their current expertise.[17] Consequently, such a scenario poses a risk of inconsistent and inadequate training.

POTENTIAL ROLE OF ARTIFICIAL INTELLIGENCE

Artificial intelligence impacts major directions and trends in healthcare, including population health, precision medicine, value-based care and translational research.[18,19] The emergence of chatbots, particularly those powered by advanced large language models (LLMs) like GPT-4 and Google BARD, has sparked renewed discussions about the integration of AI into medical education, hospital operations and the broader healthcare ecosystem.[20,21] Various AI-driven initiatives have been implemented in clinical and educational settings, leading to notable improvements in the efficiency, accuracy and overall quality of healthcare services.[16,22,23,24,25,26]

We propose two future models to illustrate how medical education and physician training could be promoted in the era of AI. The future ‘AI-assisted model’ [Figure 1b] provides more effective teaching, and a higher ‘density’ of knowledge is imparted at each stage. In this model, medical students could potentially attain a comparable ‘level’ of basic medical knowledge to that of residents in the traditional model. Similarly, residents could achieve an advanced knowledge level on a par with consultants in the traditional framework. We envision that consultants will continually acquire new knowledge and stay updated with the latest clinical practice/guidelines with the support of AI. Furthermore, the future ‘AI-integrated’ model [Figure 1c] will also help to accelerate the acquisition of medical knowledge, integrate basic and clinical training phases, shorten the training duration in medical schools, and democratise medical knowledge, ensuring each physician has equal access to advanced medical knowledge, even post-residency training.

In the subsequent sections, we delve further into this topic, offering specific examples and in-depth discussions regarding the utilisation of AI in different medical training stages.

AI-integrated preclinical medical education

The future of preclinical education is envisioned to be student-centred, complemented by an up-to-date curriculum and abundant teaching resources at a low cost. These approaches will allow students to receive more tailored knowledge that addresses their individual learning needs, enhances their studying efficiency and fosters problem-solving skills.[27,28,29] The utilisation of LLM chatbots could potentially aid in future preclinical education by offering rapid information retrieval and systematic integration while being user-friendly and cost-effective.

We explored multiple scenarios where medical students could leverage on LLMs to augment their learning. For instance, when students encounter a new subject, it is essential to extract and acquire foundational information and knowledge. We explored the use of the ‘Scholar AI’ plugin in ChatGPT-4 to facilitate this learning process[30] [Table S1, Supplemental Digital Appendix]. This plugin is capable of generating notes and summarising key points from these papers, proving to be particularly useful when focusing on specific aspects of a disease [Table S1]. However, it is important to note that the plugin has limitations, in that it is confined to open-access articles by Springer Nature, thereby restricting its access to a broader range of papers.

IMind mapping is a practical approach to cultivating students’ logical thinking skills. To further enhance this function, the ‘Show Me Diagrams’ plugin embedded in ChatGPT-4[31] can produce diagrams to illustrate the causes related to symptoms. Such visual aids assist students in systematically exploring relevant knowledge, facilitating a holistic approach to differential diagnosis.

Notably, LLMs have achieved the equivalence of a passing score on the United States Medical Licensing Examination (USMLE).[32] In this regard, LLMs can be a useful learning tool to aid in revision. For instance, as demonstrated in Table S2 [see Supplemental Digital Appendix], when ChatGPT-4 was tasked with generating questions on ‘ischaemic preconditioning’, it produced well-structured questions that inspect students’ comprehension on the topic. Thus, LLMs may be valuable in generating topic-specific questions for efficient self-assessments and lesson reviews. However, it should be noted that, while LLMs can generate USMLE-styled questions, the quality of their responses may vary considerably and warrant cautious usage. For example, when we prompted ChatGPT-4 to generate a batch of questions related to broad topics such as ‘cardiovascular and circulation systems’, it returned low-quality questions or even incorrect answers [Table S2].

AI-integrated resident training

In many countries’ residency programmes, there is an increased focus on scientific research training. LLMs present multifaceted advantages for students and residents in the realm of scientific research. First, LLMs can greatly assist students in efficiently identifying knowledge gaps and formulating innovative research questions.[33] By providing students with comprehensive literature reviews, LLMs enable critical evaluation of existing research findings.[21] Second, LLMs can support students in the revision and editing of manuscripts and can format manuscripts automatically according to guidelines of targeted academic journals, thus enhancing productivity.[21] Moreover, LLMs can provide coding assistance for statistical analysis, ensuring robust and valid findings from research.[34]

In addition, the natural language processing ability of LLMs can facilitate efficient management of administrative tasks for residents. By inputting specific patient details and emphasising key concepts, LLMs can swiftly generate detailed discharge summaries. Through automated text drafting, LLMs could generate comprehensive discharge summaries within seconds after entering specific patient details and emphasising key concepts.[35] Moreover, integrating medical-related plugins into LLMs can enhance information retrieval and access to guidelines, and improve the accuracy and comprehensiveness of medical record notes. The continuous learning and error detection abilities of AI systems are expected to produce more accurate and uniform documentation. This will enable residents to generate high-quality medical records efficiently, freeing up more time for direct patient care.

Traditional residency training with standardised patients (SPs) has been shown to be costly and time-consuming.[36] As telemedicine visits have become more prevalent in clinical practice, many institutions have also started incorporating virtual SP encounters into their communication training modules.[37] Simulation-based learning that involves interactions with SPs offers a safe and effective method for residents to practise clinical skills before engaging real patients.[38] In this context, we also explored the use of LLMs to simulate patient cases and interactions with patients. Using ChatGPT-4, we inputted the medical history of a patient with diabetic nephropathy (with modifications to ensure privacy) and subsequently prompted ChatGPT-4 to act as an SP. ChatGPT-4 was found to be able to generate paired questions and answers of high quality [Figure S1, Supplemental Digital Appendix]. This approach will allow for rapid generation of diverse virtual SPs to suit different training needs. Moreover, when integrated with virtual reality technology, further visual cues and auditory communication can be blended to enrich the simulation and learning experience, potentially surpassing the current pedagogy of traditional SP-based curricula.[38] ChatGPT-4 can also directly generate simulated medical records for specific diseases, allowing residents to engage with a variety of cases without infringing patient privacy [Figure 2]. Taken together, these advanced approaches will help to enhance residents’ real-world clinical experience and learning efficiency [Figure 1c].

Figure 2.

Figure 2

Open in a new tab

ChatGPT4-simulated case of coronary heart disease.

AI-integrated training for attending physician/consultant

For those early in their attending careers, accurate clinical decision-making is critical. The integration of AI into clinical practice may help to enhance physicians’ capabilities in different facets [Figure 1c]. First, AI-powered tools can improve diagnostic accuracy, especially in the context of complex medical conditions.[39] By leveraging AI algorithms, physicians can conduct in-depth risk analyses, formulating individualised treatment plans for patients. In addition, the capability of AI to detect and flag errors in clinical recordings or decisions could help to enhance patient safety.[40] Furthermore, by drawing insights from AI-assisted clinical processes, attending physicians can continually refine their clinical expertise acumen. Overall, appropriate integration of AI into clinical practice can help in the continuous professional growth of attending physicians.

AI-integrated future clinic

Generalist Medical AI (GMAI) represents a suite of foundation models proficient in integrating multimodal data, encompassing medical history text, imaging data and laboratory findings. These models can process and integrate information, and provide lesion visualisation and decision support.[41] Envisioning the future clinical landscape, when GMAI is synergised with portable devices such as smart eyewear/goggles, it could facilitate real-time health monitoring.[41] Leveraging GMAI in tandem with its complementary portable device, physicians can effortlessly transcribe clinical notes during patient consultations while concurrently receiving real-time insights on relevant medical history inquiries or recommended physical examinations.[19] Consequently, AI-integrated medicine could enhance precision medicine and personalised interventions. This new paradigm not only augments physicians’ clinical problem-solving ability, but also paves the way for innovative solutions to intricate clinical challenges.[42]

ARTIFICIAL INTELLIGENCE AS AN ACADEMIC SUBJECT

Despite advancements in technological innovation, the incorporation of AI into healthcare is still faced with substantial challenges. There is a pressing need for experts who are well versed in both clinical care intricacies and the transformative capabilities of AI to bridge the gap between AI development and its clinical implementation.

Current state of AI curriculum in medical education

Surveys conducted among medical students and faculties in USA, UK, Germany and Turkey regarding the perceptions of AI in medicine revealed that the majority of surveyed participants opined that AI has the potential to revolutionise medical practice and enhance physicians’ access to information. In addition, in another survey among students, the majority concurred that AI can streamline patient access to healthcare services and improve healthcare quality.[43,44,45,46] Moreover, two review papers collectively emphasise the necessity of adapting medical curricula in response to AI advancements.[47,48] However, previous surveys also brought attention to emerging concerns regarding potential pitfalls associated with AI, including data biases, ethical dilemmas and the occasional AI ‘hallucination’ — all of which could erode trust between physicians and patients. In addition, current studies provide little consensus on the content and pedagogical approaches for AI courses in medical education, highlighting a gap in providing clear guidelines for practical application.[43,47,48,49,50] Hence, it is essential to make a pivotal exploration in medical education and equip future physicians with the knowledge and skills to navigate this digital era.[51]

The World Medical Association advocates the need to update medical curricula to ensure healthcare professionals are well versed in AI’s benefits and risks in healthcare.[52] In line with this, a significant majority of the surveyed medical students and doctors from countries like USA, UK, Germany and Turkey have expressed their approval of incorporating structured AI-related training courses into medical curricula.[43,44,45,53] However, the development of a universally endorsed AI curriculum has remained elusive. This underscores the urgency for a standardised AI curriculum introduced early in medical education. In the following section, we outline the tiers of AI proficiency that ought to be integrated into a tailored AI curriculum for medical training.

Stages of AI expertise in the AI curriculum

To tailor the AI proficiency to the varying stages of a medical professional’s career, we propose categorising roles into ‘general AI users’, ‘medicine-AI facilitators’ and ‘AI innovators’ [Figure 3]. Correspondingly, these roles demand varying degrees of AI expertise: ‘foundational AI literacy’, ‘intermediate knowledge’ and ‘mastery expertise’ [Figure 3].

Figure 3.

Figure 3

Open in a new tab

Diagram shows the categorisation of artificial intelligence (AI) curricula professionals — general AI users, medicine-AI facilitators and AI innovators.

General users of AI: Acquiring the ‘essentials’ of AI

The ‘general AI users’ category may encompass physicians, nurses and other allied health staffs. The primary objective for this group is to integrate basic AI knowledge into clinical practice, leveraging AI technologies to enhance healthcare services [Table 1]. At this stage, it is imperative for these users to acquire foundational AI literacy. Moreover, it is crucial for these users to evaluate the validity of AI’s performance, understanding its benefits, limitations and potential risks.

Table 1.

Roles and competencies across different tiers of artificial intelligence (AI) expertise in healthcare.

Curriculum objective General AI users Medicine-AI facilitators AI innovators
Knowledge and skills Grasp of fundamental AI concepts 1. Having basic coding skills 2. Intermediate understanding of AI architecture and deep learning schemata 1. Possess expertise in creating and developing innovative AI algorithms
2. Spearheading foundational AI research
Assessment capability Understand the reliability and limitations of AI outputs Interpret parameters to assess AI performance 1. Mastery in developing advanced and robust validation techniques
2. Ensure AI systems meet the highest standards and optimise for clinical utility
Application and integration 1. Basic use of AI tools in clinical settings
2. Navigate evolving AI products in healthcare		 1. Drive research on AI’s clinical application
2. Facilitate integration of AI tools into clinical workflows		 1. Innovate by designing and developing novel AI applications tailored for specific clinical needs
2. Drive implementation and integration of AI solutions into health systems
Ethical and legal awareness 1. Basic understanding of AI’s ethical and legal implications
2. Recognise AI’s credibility, benefits and risks		 1. In-depth understanding of AI ethics and legislation
2. Capable of participating in and facilitating the development of AI-clinical frameworks		 1. Lead discussions on complex AI ethics topics
2. Lead the establishment of standards and guidelines for AI’s clinical integration
Open in a new tab

Medicine-AI facilitators: Delving into the ‘intermediate’ aspects of AI

‘Medicine-AI facilitators’ are often academic clinicians or clinician-scientists. These roles will act as ‘conduits’ between the realms of medicine and technology. They are expected to possess a good grasp of AI architecture, deep learning model frameworks and basic coding skills, and should be able to interpret and validate AI models’ performance. By mastering these aspects, these facilitators can actively engage and drive research related to the clinical application of AI. Being equipped with knowledge of AI ethics and legislation, they will also be well positioned to establish clinical frameworks that safeguard patients’ privacy and data safety.

AI innovators: Achieving mastery in AI expertise

These innovators are ‘clinician-computer scientists’ with specialisation in both AI and healthcare. They are at the forefront of designing and developing innovative AI algorithms, and spearheading ground-breaking AI research. With a profound understanding of cutting-edge advancements in AI, these innovators can leverage the latest AI techniques and solutions to address unique clinical gaps.[54] With a comprehensive view that spans the technical, economic and ethical challenges of AI’s clinical implementation, they are uniquely qualified to harmonise these diverse aspects and develop fit-for-purpose innovation.[55,56,57] In addition, they can lead the establishment of standards and guidelines for AI’s clinical integration.[58]

In summary, AI curriculum should continuously adapt to the diverse needs of healthcare systems. Its core principle remains empowering individuals to leverage AI for better clinical outcomes.[49] As AI becomes central to medicine, a tiered curriculum tailored for medical education can enhance both training and educational standards. Collaboration among ‘general AI users’, ‘medicine-AI facilitators’ and ‘AI innovators’ is key to fostering an AI-integrated medical education.

CHALLENGES IN INTEGRATING AI INTO MEDICAL EDUCATION

The remarkable potential of AI is accompanied by significant challenges that necessitate careful consideration. One such challenge arises from the black box operation of LLMs, leading to the potential generation of false or offensive outputs.[54] In addition, the AI systems pose risks to patient privacy and security, particularly when sensitive information is embedded in the medical data used to train these models. Inaccurate or incomplete medical treatment information can also have severe consequences. To date, there is no established framework for accountability related to these implications.[20,42] Transparency and patients’ acceptability should also be considered when deploying AI solutions.[33] Furthermore, the use of AI could potentially lead to over-reliance, stifling independent thinking and potentially overlooking AI’s inherent risk.[59] Currently, majority of the students may still lack awareness of AI risks. In a survey conducted in universities and hospitals in China, only 6.0% of students felt competent enough to inform patients about the features and risks of AI.[60] Hence, physicians need to remain at the helm of clinical decision-making, discerning the appropriate contexts for AI, interpreting its results against established accuracy benchmarks and staying vigilant about potential biases and limitations.

FUTURE AREAS AND CONCLUSION

Leveraging on AI as a learning tool allows students to expand their knowledge effectively and stay up-to-date. Evaluating vast amounts of rapidly changing information during this process can facilitate the development of critical thinking, a crucial soft skill in a clinical setting.[61] Early integration of an AI curriculum can amplify these benefits.

Concerted efforts are needed for the practical implementation of AI curriculum into the medical education system. First, effective teaching with a higher density of knowledge imparted at each stage is expected to create more time and space for AI-related learning. Second, prioritising the importance of AI education requires reallocation of resources within the education system. Third, multidisciplinary collaborations among experts are expected to identify core AI curricula tailored specifically for medical students. Open-access databases of these exquisite courses and structured textbooks could serve as bases for the widespread implementation of AI curricula. However, in addition to the classical AI curriculum, more flexible teaching forms addressing more professional applications are required to cope with the rapid evolution of AI knowledge. For instance, case-based workshops such as proper application of an AI assisted-glaucoma diagnosis system, which emphasises experiential learning, could accelerate mastery of a narrow clinical application of AI.[62,63,64]

The success of introducing AI in the proposed manner can be measured through various metrics such as improved efficiency in assessment processes, enhanced accuracy of evaluations and the ability to provide personalised, adaptive assessments that cater to individual learning needs. Although vast randomised controlled trials have been launched in clinics to assess AI application in improving doctors’ clinical performance and have shown promising results, only quantitative indices such as diagnostic accuracy are being studied.[65] The long-term effects of AI integration and other humanistic indicators, for example, empathy and comprehensiveness, were not included.[66] Therefore, a holistic profiling of medical professionals is essential to advance AI-integrated medical education, and AI may facilitate the assessment based on AI-assisted dataset collection and AI-generated analytics. Prober and Desai[67] proposed a systemic evaluation approach that encompasses academic proficiency, adaptive learning skills, inquisitiveness, emotional maturity, empathy and communication skills, rather than standard test scores. The transformative impact of AI on assessments and examinations is expected to streamline and personalise the assessment process, which can increase efficiency, personalisation and data-driven insights. Furthermore, AI can assist in the analysis of large-scale data, leading to more comprehensive insights into student performance and learning trends. Integrating AI into assessment can allow better customisation of content and the difficulty level of assessment, tailoring the assessment to the specific competencies and learning needs of each student.

The impact of AI on medical and physician education is complex and multifaceted. The ability of AI to provide access to a wide range of educational resources and tools, regardless of geographical location, could potentially improve access to high-quality medical education, especially for individuals in remote or underserved areas. However, there may still be disparities in the training and familiarity with AI tools across medical institutions or healthcare systems in different countries, leading to unequal opportunities for students and physicians.[68,69]

Continuous efforts to integrate AI education into medical curricula and continuing medical education programmes may help ensure that all healthcare professionals are well equipped and ready to leverage AI-related resources. Nevertheless, the extent to which AI exacerbates or narrows the inequities in medical and physician education will depend on how effectively stakeholders address these challenges and prioritise equitable access and applicability.

In conclusion, AI holds promise in enhancing current medical education and training frameworks, potentially leading to a revolutionary shift from traditional pedagogy to AI-assisted or AI-integrated pedagogy models. As we stand on the cusp of this transformative era, medical educators must be actively involved in this development and adapt to these changes. This would help to ensure that future physicians are thoroughly prepared for the evolving challenges in healthcare.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Supplemental digital content

Appendix at http://links.lww.com/SGMJ/A103

APPENDIX

Table S1.

Demonstration on the use of Scholar AI plugin in ChatGPT for extraction and summary of top cited diabetic peripheral neuropathy papers.

Prompt: please find 5 top cited papers about diabetic peripheral neuropathy (DPN).
1. Title: Diabetic peripheral neuropathy: advances in diagnosis and strategies for screening and
early intervention.
• Authors: D. Selvarajah, D. Kar, K. Khunti, M. Davies, A. Scott, Jeremy Walker, S. Tesfaye
• Publication Date: 2019-10-14
• Cited by: 164
• Abstract: This paper reviews innovative point-of-care devices that enable the early diagnosis of DPN and assesses the evidence for early risk factor-based management strategies to reduce the incidence and slow the progression of DPN. It also proposes a framework for screening and early multifactorial interventions as the best prospect for preventing or halting DPN and its devastating sequelae.
• URL: https://dx.doi.org/10.1016/s2213-8587(19)30081-6
2. Title: The risk factors for diabetic peripheral neuropathy: A meta-analysis
• Authors: Xiu-juan Liu, Yu-yan Xu, M. An, Q. Zeng
• Publication Date: 2019-02-20
• Cited by: 90
• Abstract: This study systematically evaluated DPN’s influencing factors in patients with type 2 diabetes using evidence-based medicine. The results suggested that the duration of diabetes, age, HbA1c, and DR are associated with significantly increased risks of DPN among diabetic patients, while BMI, smoking, TG, and TC did not indicate any risks of increasing DPN.
• URL: https://dx.doi.org/10.1371/journal.pone.0212574
3. Title: Risk Factors for Diabetic Peripheral Neuropathy and Cardiovascular Autonomic Neuropathy in the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study
• Authors: B. Braffett, R. Gubitosi-Klug, J. Albers, E. Feldman, Catherine L. Martin, N. White, T. Orchard, M. Lopes-Virella, J. Lachin, R. Pop-Busui
• Publication Date: 2020-02-12
• Cited by: 80
• Abstract: This study evaluated multiple risk factors and phenotypes associated with DPN and CAN in a large cohort of participants with type 1 diabetes, followed for >23 years. The findings identify risk factors and phenotypes of participants with diabetic neuropathy that can be used in the design of new interventional trials and for personalized approaches to neuropathy prevention.
• URL: https://dx.doi.org/10.2337/db19-1046
4. Title: Measuring Plantar Tissue Stress in People With Diabetic Peripheral Neuropathy: A
Critical Concept in Diabetic Foot Management
• Authors: P. Lazzarini, R. Crews, J. V. van Netten, S. Bus, M. Fernando, P. Chadwick, B. Najafi
• Publication Date: 2019-04-29
• Cited by: 70
• Abstract: This article reviews the concept of PTS, the mechanical factors involved, and the findings of pivotal articles reporting measures of PTS in people with neuropathy. It also proposes recommended clinical and research standards for measuring and reporting PTS in people with neuropathy and forecasts future clinical, research, and technological advancements that may use PTS to highlight the importance of this critical concept in the prevention and management of diabetic foot ulcers.
• URL: https://dx.doi.org/10.1177/1932296819849092
5. Title: Structural and Functional Abnormalities of the Primary Somatosensory Cortex in Diabetic Peripheral Neuropathy: A Multimodal MRI Study
• Authors: D. Selvarajah, I. Wilkinson, Fang Fang, A. Sankar, J. Davies, Elaine Boland, J. Harding, G. Rao, R. Gandhi, I. Tracey, S. Tesfaye
• Publication Date: 2019-01-07
• Cited by: 51 times
• Abstract: This study examined whether changes in brain volume alter the functional organization of the somatosensory cortex and how this relates to the various painful DSP clinical phenotypes. The study found significant anatomical and functional changes in the somatosensory cortex. Subjects with painful DSP insensate have the lowest somatosensory cortical thickness, with expansion of the area representing pain in the lower limb to include face and lip regions. There was a significant relationship between anatomical and functional changes within the somatosensory cortex and severity of the peripheral neuropathy.
• URL: https://dx.doi.org/10.2337/db18-0509 Prompt: please summarize the risk factors of diabetic peripheral neuropathy (DPN). A summary of the risk factors for DPN based on the paper titled “The risk factors for diabetic peripheral neuropathy: A meta-analysis”
Prompt: please summarize the risk factors of diabetic peripheral neuropathy (DPN).
A summary of the risk factors for DPN based on the paper titled “The risk factors for diabetic peripheral neuropathy: A meta-analysis”
 • Duration of Diabetes: The longer the duration of diabetes, the higher the risk of developing DPN
 • Age: Older age is associated with an increased risk of DPN.
 • Glycosylated Hemoglobin A1c (HbA1c): Elevated levels of HbA1c, which indicates poor blood sugar control, are linked to a higher risk of DPN.
 • Diabetic Retinopathy (DR): The presence of diabetic retinopathy, a complication of diabetes that affects the eyes, is associated with an increased risk of DPN.
 • Smoking: While some studies suggest smoking as a risk factor, the paper’s meta-analysis did not find a significant association between smoking and an increased risk of DPN.
 • Body Mass Index (BMI): Similarly, while some studies have pointed to BMI as a potential risk factor, the meta-analysis did not find a significant link between BMI and DPN risk.
 • Other Factors: The paper mentions other potential risk factors like fasting plasma glucose (FPG), blood urea nitrogen (BUN), diastolic blood pressure (DBP), among others. However, the strength of the association between these factors and DPN varies across studies.
It’s important to note that while some factors like the duration of diabetes, age, HbA1c, and DR showed a strong association with DPN, others like BMI and smoking did not indicate any significant risks in the meta-analysis. The paper emphasizes the need for high-quality prospective cohort studies to further validate these findings.
Open in a new tab

Table S2.

Examples of ChatGPT-4 generated USMLE-style questions and model answers.

Prompt: Can you generate USMLE-style questions about Ischemic preconditioning.
2 examples of ChatGPT-4 generated questions:
Question: A 58-year-old man with a history of type 2 diabetes and hypertension comes to the clinic complaining of intermittent chest pain for the past two months. He describes the pain as a tight squeezing sensation that usually lasts for 1-2 minutes and resolves spontaneously. He also notes that these episodes are triggered by physical exertion and are relieved with rest. Recently, he experienced a more prolonged episode of similar pain at rest, which resolved after 10 minutes. An ECG during one of the brief episodes of pain revealed transient ST-segment elevations, which normalized after the pain subsided. The physician is concerned that these transient episodes of ischemia have protected his myocardium during the prolonged episode. This protective phenomenon is best described by which of the following mechanisms?
A.Reperfusion injury
B.Collateral circulation development
C.Hibernating myocardium
D.Ischemic preconditioning
E.Myocardial stunning
Answer: D. Ischemic preconditioning
Explanation:
Ischemic preconditioning refers to the phenomenon where brief, intermittent episodes of sublethal ischemia can protect the myocardium from a subsequent, more prolonged ischemic insult. This is believed to be due to a complex interplay of intracellular signaling pathways that are activated during the short ischemic episodes, which then render the myocardial cells more resistant to injury during the subsequent prolonged ischemic event. The transient ST-segment elevations during the brief episodes of chest pain indicate episodes of sublethal ischemia, which might have provided a protective effect
during his more prolonged episode of ischemic pain.
Question: A 65-year-old man with a history of coronary artery disease is scheduled for a coronary artery bypass graft (CABG) surgery. The surgeon plans to use ischemic preconditioning as a protective measure. Which of the following best describes the mechanism of ischemic preconditioning?
A.Induction of hypothermia
B.Administration of antiplatelet agents
C.Brief episodes of ischemia and reperfusion
D.Continuous infusion of nitrates
E.Prolonged total ischemia
Answer: C. Brief episodes of ischemia and reperfusion
Explanation: Ischemic preconditioning involves subjecting the heart to brief, non-lethal episodes of ischemia followed by reperfusion. This process activates endogenous protective mechanisms that reduce the extent of myocardial injury during subsequent prolonged ischemia.
Prompt: Can you generate USMLE-style questions of the cardiovascular and circulation system.
1 example of ChatGTP-4 generated question:
Question: Which of the following factors primarily determines myocardial oxygen demand?
A)Heart rate
B)Venous return
C)Preload
D)Afterload
E)Wall tension
Answer: E) Wall tension
Explanation: Wall tension, influenced by factors like preload, afterload and heart rate, is the primary determinant of myocardial oxygen demand.
Open in a new tab

Figure S1.

Figure S1

Open in a new tab

Demonstration of using ChatGPT-4 to simulate standardised patient encounters for diabetic nephropathy.

REFERENCES

  • 1.Flexner A. Medical Education in the United States and Canada: A Report to the Carnegie Foundation for the Advancement of Teaching. The Carnegie Foundation for the Advancement of Teaching. 1910 [PMC free article] [PubMed] [Google Scholar]
  • 2.Frenk J, Chen L, Bhutta ZA, Cohen J, Crisp N, Evans T, et al. Health professionals for a new century: Transforming education to strengthen health systems in an interdependent world. Lancet. 2010;376:1923–58. doi: 10.1016/S0140-6736(10)61854-5. [DOI] [PubMed] [Google Scholar]
  • 3.Wang C, Chen S, Zhu J, Li W. China’s new 4+4 medical education programme. Lancet. 2019;394:1121–3. doi: 10.1016/S0140-6736(19)32178-6. [DOI] [PubMed] [Google Scholar]
  • 4.Wang W. Medical education in china: Progress in the past 70 years and a vision for the future. BMC Med Educ. 2021;21:453. doi: 10.1186/s12909-021-02875-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Frenk J, Chen LC, Chandran L, Groff EOH, King R, Meleis A, et al. Challenges and opportunities for educating health professionals after the covid-19 pandemic. Lancet. 2022;400:1539–56. doi: 10.1016/S0140-6736(22)02092-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wu Q, Zhao L, Ye X-C. Shortage of healthcare professionals in china. BMJ. 2016;354:i4860. doi: 10.1136/bmj.i4860. [DOI] [PubMed] [Google Scholar]
  • 7.Measuring the availability of human resources for health and its relationship to universal health coverage for 204 countries and territories from 1990 to 2019: A systematic analysis for the global burden of disease study 2019. Lancet. 2022;399:2129–54. doi: 10.1016/S0140-6736(22)00532-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hu X, Du B. 4+4 medical education: A word of caution. Lancet. 2020;395:688. doi: 10.1016/S0140-6736(19)32996-4. [DOI] [PubMed] [Google Scholar]
  • 9.Campbell J, Dussault G, Buchan J, Pozo-Martin F, Arias MG, Leone C, et al. WHO Press; 2013. A universal truth: No health without a workforce. [Google Scholar]
  • 10.Blood AD, Farnan JM, Fitz-William W. Curriculum changes and trends 2010-2020: A focused national review using the aamc curriculum inventory and the lcme annual medical school questionnaire part ii. Acad Med. 2020;95:S5–14. doi: 10.1097/ACM.0000000000003484. [DOI] [PubMed] [Google Scholar]
  • 11.Huggett KN, Quesnelle KM, Jeffries MB, editors. Switzerland: Springer Cham; 2022. An introduction to medical teaching. [Google Scholar]
  • 12.Asch DA, Grischkan J, Nicholson S. The cost, price, and debt of medical education. N Engl J Med. 2020;383:6–9. doi: 10.1056/NEJMp1916528. [DOI] [PubMed] [Google Scholar]
  • 13.Zhu J, Li W, Chen L. Doctors in china: Improving quality through modernisation of residency education. Lancet. 2016;388:1922–9. doi: 10.1016/S0140-6736(16)00582-1. [DOI] [PubMed] [Google Scholar]
  • 14.Xiao YT, Wang YT. Standardised residency training: Students’concerns. Lancet. 2017;389:905. doi: 10.1016/S0140-6736(17)30596-2. [DOI] [PubMed] [Google Scholar]
  • 15.Finn KM, Metlay JP, Chang Y, Nagarur A, Yang S, Landrigan CP, et al. Effect of increased inpatient attending physician supervision on medical errors, patient safety, and resident education: A randomized clinical trial. JAMA Intern Med. 2018;178:952–9. doi: 10.1001/jamainternmed.2018.1244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Han ER, Yeo S, Kim MJ, Lee YH, Park KH, Roh H. Medical education trends for future physicians in the era of advanced technology and artificial intelligence: An integrative review. BMC Med Educ. 2019;19:460. doi: 10.1186/s12909-019-1891-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Forsythe RO, Suttie SA. Enhancing junior doctors’ working lives. Surgery (Oxford, Oxfordshire) 2020;38:607–11. doi: 10.1016/j.mpsur.2020.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Rajpurkar P, Chen E, Banerjee O, Topol EJ. Ai in health and medicine. Nat Med. 2022;28:31–8. doi: 10.1038/s41591-021-01614-0. [DOI] [PubMed] [Google Scholar]
  • 19.Haug CJ, Drazen JM. Artificial intelligence and machine learning in clinical medicine, 2023. N Engl J Med. 2023;388:1201–8. doi: 10.1056/NEJMra2302038. [DOI] [PubMed] [Google Scholar]
  • 20.Lee P, Bubeck S, Petro J. Benefits, limits, and risks of gpt-4 as an AI chatbot for medicine. N Engl J Med. 2023;388:1233–9. doi: 10.1056/NEJMsr2214184. [DOI] [PubMed] [Google Scholar]
  • 21.Van Dis EA, Bollen J, Zuidema W, van Rooij R, Bockting CL. Chatgpt: Five priorities for research. Nature. 2023;614:224–6. doi: 10.1038/d41586-023-00288-7. [DOI] [PubMed] [Google Scholar]
  • 22.Bresnick G, Cuadros JA, Khan M, Fleischmann S, Wolff G, Limon A, et al. Adherence to ophthalmology referral, treatment and follow-up after diabetic retinopathy screening in the primary care setting. BMJ Open Diabetes Res Care. 2020;8:e001154. doi: 10.1136/bmjdrc-2019-001154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Dong L, He W, Zhang R, Ge Z, Wang YX, Zhou J, et al. Artificial intelligence for screening of multiple retinal and optic nerve diseases. JAMA Netw Open. 2022;5:e229960. doi: 10.1001/jamanetworkopen.2022.9960. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ting DSW, Pasquale LR, Peng L, Campbell JP, Lee AY, Raman R, et al. Artificial intelligence and deep learning in ophthalmology. Br J Ophthalmol. 2019;103:167–75. doi: 10.1136/bjophthalmol-2018-313173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Institute of Medicine (US) Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: The National Academies Press; 2001. [PubMed] [Google Scholar]
  • 26.Pucchio A, Eisenhauer EA, Moraes FY. Medical students need artificial intelligence and machine learning training. Nat Biotechnol. 2021;39:388–9. doi: 10.1038/s41587-021-00846-2. [DOI] [PubMed] [Google Scholar]
  • 27.Rudolph J, Tan S, Tan S. Chatgpt: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning & Teaching. 2023;6:342–62. [Google Scholar]
  • 28.Biswas S. Chatgpt and the future of medical writing. Radiology. 2023;307:e223312. doi: 10.1148/radiol.223312. [DOI] [PubMed] [Google Scholar]
  • 29.Lee H. The rise of chatgpt: Exploring its potential in medical education. Anat Sci Educ. 2023 doi: 10.1002/ase.2270. doi: 10.1002/ase.2270. [DOI] [PubMed] [Google Scholar]
  • 30. [[Last accessed on 2023 Nov 04]]. Available at: https://scholarai.io/
  • 31.OpenWorld AI. [[Last accessed on 2023 Nov 04]]. Available at: https://openworldai.com/plugins/showme-diagrams .
  • 32.Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepaño C, et al. Performance of chatGPT on USMLE: Potential for ai-assisted medical education using large language models. PLOS Digit Health, 2023;2:e0000198. doi: 10.1371/journal.pdig.0000198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Sallam M. Chatgpt utility in healthcare education, research, and practice: Systematic review on the promising perspectives and valid concerns. Healthcare (Basel, Switzerland) 2023;11:887. doi: 10.3390/healthcare11060887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Sanmarchi F, Bucci A, Nuzzolese AG, Carullo G, Toscano F, Nante N, et al. A step-by-step researcher’s guide to the use of an AI-based transformer in epidemiology: An exploratory analysis of chatgpt using the strobe checklist for observational studies. J Public Health. 2023:1–36. doi: 10.1007/s10389-023-01936-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Patel SB, Lam K. Chatgpt: The future of discharge summaries? Lancet Digit Health. 2023;5:e107–8. doi: 10.1016/S2589-7500(23)00021-3. [DOI] [PubMed] [Google Scholar]
  • 36.Bosek MS, Li S, Hicks FD. Working with standardized patients: A primer. Int J Nurs Educ Scholarsh. 2007;4 doi: 10.2202/1548-923X.1437. doi: 10.2202/1548-923X.1437. [DOI] [PubMed] [Google Scholar]
  • 37.Suneja M, Thoma KD, Franklin E, Rowat J. Pivoting during a pandemic: Reimagining intern orientation epa-based osce. South Med J. 2023;116:312–6. doi: 10.14423/SMJ.0000000000001525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.C. Donald Combs PFC. Emerging roles of virtual patients in the age of AI. AMA J Ethics. 2019;21:E153–9. doi: 10.1001/amajethics.2019.153. [DOI] [PubMed] [Google Scholar]
  • 39.Rajpurkar P, Lungren MP. The current and future state of AI interpretation of medical images. N Engl J Med. 2023;388:1981–90. doi: 10.1056/NEJMra2301725. [DOI] [PubMed] [Google Scholar]
  • 40.Ellahham S. Artificial intelligence: The future for diabetes care. Am J Med. 2020;133:895–900. doi: 10.1016/j.amjmed.2020.03.033. [DOI] [PubMed] [Google Scholar]
  • 41.RaviChandran N, Teo Z, Ting D. Artificial intelligence enabled smart digital eye wearables. Curr Opin Ophthalmol. 2023;34:414–21. doi: 10.1097/ICU.0000000000000985. [DOI] [PubMed] [Google Scholar]
  • 42.Thirunavukarasu AJ, Ting DSJ, Elangovan K, Gutierrez L, Tan TF, Ting DSW. Large language models in medicine. Nat Med. 2023;29:1930–40. doi: 10.1038/s41591-023-02448-8. [DOI] [PubMed] [Google Scholar]
  • 43.Civaner MM, Uncu Y, Bulut F, Chalil EG, Tatli A. Artificial intelligence in medical education: A cross-sectional needs assessment. BMC Med Educ. 2022;22:772. doi: 10.1186/s12909-022-03852-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Wood EA, Ange BL, Miller DD. Are we ready to integrate artificial intelligence literacy into medical school curriculum: Students and faculty survey. J Med Educ Curric Dev. 2021;8 doi: 10.1177/23821205211024078. 23821205211024078. doi: 10.1177/23821205211024078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Bisdas S, Topriceanu CC, Zakrzewska Z, Irimia AV, Shakallis L, Subhash J, et al. Artificial intelligence in medicine: A multinational multi-center survey on the medical and dental students’perception. Front Public Health. 2021;9:795284. doi: 10.3389/fpubh.2021.795284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Pinto Dos Santos D, Giese D, Brodehl S, Chon SH, Staab W, Kleinert R, et al. Medical students’ attitude towards artificial intelligence: A multicentre survey. Eur Radiol. 2019;29:1640–6. doi: 10.1007/s00330-018-5601-1. [DOI] [PubMed] [Google Scholar]
  • 47.Grunhut J, Wyatt AT, Marques O. Educating future physicians in artificial intelligence (AI): An integrative review and proposed changes. J Med Educ Curric Dev. 2021;8 doi: 10.1177/23821205211036836. 23821205211036836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Lee J, Wu AS, Li D, Kulasegaram KM. Artificial intelligence in undergraduate medical education: A scoping review. Acad Med. 2021;96:S62–70. doi: 10.1097/ACM.0000000000004291. [DOI] [PubMed] [Google Scholar]
  • 49.Cooper A, Rodman A. Ai and medical education-A 21st-century pandora’s box. N Engl J Med. 2023;389:385–7. doi: 10.1056/NEJMp2304993. [DOI] [PubMed] [Google Scholar]
  • 50.Beam AL, Drazen JM, Kohane IS, Leong TY, Manrai AK, Rubin EJ. Artificial intelligence in medicine. N Engl J Med. 2023;388:1220–1. doi: 10.1056/NEJMe2206291. [DOI] [PubMed] [Google Scholar]
  • 51.Wartman SA, Combs CD. Medical education must move from the information age to the age of artificial intelligence. Acad Med. 2018;93:1107–9. doi: 10.1097/ACM.0000000000002044. [DOI] [PubMed] [Google Scholar]
  • 52.OECD AI. Statement on Augmented Intelligence in Medical Care. [[Last accessed on November 4, 2023]]. Available at: https://oecd.ai/en/catalogue/tools/statement-on-augmented-intelligence-in-medical-care .
  • 53.Sit C, Srinivasan R, Amlani A, Muthuswamy K, Azam A, Monzon L, et al. Attitudes and perceptions of UK medical students towards artificial intelligence and radiology: A multicentre survey. Insights Imaging. 2020;11:14. doi: 10.1186/s13244-019-0830-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Deng J, Lin Y. The benefits and challenges of chatGPT: An overview. FCIS. 2023;2:81–3. [Google Scholar]
  • 55.Abookire S, Plover C, Frasso R, Ku B. Health design thinking: An innovative approach in public health to defining problems and finding solutions. Front Public Health. 2020;8:459. doi: 10.3389/fpubh.2020.00459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Wolcott MD, McLaughlin JE, Hubbard DK, Rider TR, Umstead K. Twelve tips to stimulate creative problem-solving with design thinking. Med Teach. 2021;43:501–8. doi: 10.1080/0142159X.2020.1807483. [DOI] [PubMed] [Google Scholar]
  • 57.Tejani AS. Identifying and addressing barriers to an artificial intelligence curriculum. J Am Coll Radiol. 2021;18:605–7. doi: 10.1016/j.jacr.2020.10.001. [DOI] [PubMed] [Google Scholar]
  • 58.Solomonides AE, Koski E, Atabaki SM, Weinberg S, McGreevey JD, Kannry JL, et al. Defining AMIA’s artificial intelligence principles. J Am Med Inform Assoc. 2022;29:585–91. doi: 10.1093/jamia/ocac006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Passi S, Vorvoreanu M. Overreliance on AI: Literature review. [[Last accessed on 2023 Nov 04]]. Available at: https://www.microsoft.com/en-us/research/publication/overreliance-on-ai-literature-review/
  • 60.Ahmad SF, Han H, Alam MM, Rehmat MK, Irshad M, Arraño-Muñoz M, et al. Impact of artificial intelligence on human loss in decision making, laziness and safety in education. Humanit Soc Sci Commun. 2023;10:311. doi: 10.1057/s41599-023-01787-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Royce CS, Hayes MM, Schwartzstein RM. Teaching critical thinking: A case for instruction in cognitive biases to reduce diagnostic errors and improve patient safety. Acad Med. 2019;94:187–94. doi: 10.1097/ACM.0000000000002518. [DOI] [PubMed] [Google Scholar]
  • 62.Stokel-Walker C, Van Noorden R. What chatgpt and generative AI mean for science. Nature. 2023;614:214–6. doi: 10.1038/d41586-023-00340-6. [DOI] [PubMed] [Google Scholar]
  • 63.van Dis EAM, Bollen J, Zuidema W, van Rooij R, Bockting CL. ChatGPT: Five priorities for research. Nature. 2023;614:224–226. doi: 10.1038/d41586-023-00288-7. [DOI] [PubMed] [Google Scholar]
  • 64.Azamfirei R, Kudchadkar SR, Fackler J. Large language models and the perils of their hallucinations. Crit Care. 2023;27:120. doi: 10.1186/s13054-023-04393-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Xu H, Tang RSY, Lam TYT, Zhao G, Lau JYW, Liu Y, et al. Artificial intelligence-assisted colonoscopy for colorectal cancer screening: A multicenter randomized controlled trial. Clin Gastroenterol Hepatol. 2023;21:337–46.e3. doi: 10.1016/j.cgh.2022.07.006. [DOI] [PubMed] [Google Scholar]
  • 66.Krive J, Isola M, Chang L, Patel T, Anderson M, Sreedhar R. Grounded in reality: Artificial intelligence in medical education. JAMIA Open. 2023;6:ooad037. doi: 10.1093/jamiaopen/ooad037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Prober CG, Desai SV. Medical school admissions: Focusing on producing a physician workforce that addresses the needs of the united states. Acad Med. 2023;98:983–6. doi: 10.1097/ACM.0000000000005262. [DOI] [PubMed] [Google Scholar]
  • 68.The L. AI in medicine: Creating a safe and equitable future. Lancet. 2023;402:503. doi: 10.1016/S0140-6736(23)01668-9. [DOI] [PubMed] [Google Scholar]
  • 69.Hswen Y, Voelker R. New AI tools must have health equity in their DNA. JAMA. 2023;330:1604–7. doi: 10.1001/jama.2023.19293. [DOI] [PubMed] [Google Scholar]

Articles from Singapore Medical Journal are provided here courtesy of Wolters Kluwer -- Medknow Publications

RESOURCES