Artificial Intelligence in Undergraduate Medical Education: A Scoping Review

In the citing paper

“Current advances in AI have significantly influenced diagnostic and clinical decision-making [11, 12]”

In the citing paper

“This review summarized key themes regarding AI training in UGME and identified priorities to advance curricular development.”

In the citing paper

“Lee et al. carried out a scoping review of the literature from 2000 to July 2020, identifying 22 studies on AI in UGME [5].”

In the citing paper

“a lack of consensus on core competencies [5, 174, 175]”

In the citing paper

“Challenges related to faculty members, such as the lack of AI knowledge and expertise [56, 180], resistance to adopting new technologies [5, 77, 180], and struggles with staff shortages, were also acknowledged in several studies [174].”

In the citing paper

“Since that study, which included 22 publications, the number of publications on AI in UGME has increased dramatically, necessitating another examination of the AI landscape in UGME.”

In the citing paper

“there were a small number that focused on AIHEd in specific disciplines or with particular groups of participants, for example Health & Welfare (n = 14), STEM (n = 4), online or blended learning (n = 5), foreign language learning (n = 2), pre-service teachers (Salas-Pilco et al., 2022), students with disabilities (Fichten et al., 2021), and undergraduate students (Lee et al., 2021).”

In the citing paper

“According to Ouyang et al. (2020), there is a disconnect between AI technology and existing educational systems, and suggest the need for more unified, standardised frameworks that incorporate ethical principles and advocate for the development of multidisciplinary teams (Charow et al., Lee et al., 2021), with a stronger focus on more robust and ethically aware AI curricula (e.g., Grunhut et al.,”

In the citing paper

“At the same time, it has been recognised that a lack of AI literacy could lead to a shift in authority moving decision-making from clinicians to AI systems (Lee et al., 2021).”

In the citing paper

“but these syntheses largely describe broad “AI literacy” themes or program-level content rather than systematically extracting and organizing competency- and learning-objective-level outcomes for undergraduate medical education (UME).”

In the citing paper

“Several scoping reviews have examined AI within medical education23–28”

In the citing paper

“Within undergraduate medical education, Lee et al. and Simoni et al. found that most publications describe general content areas rather than detailed competency-based outcomes23,31.”

In the citing paper

“Despite the support for including AI education, financial barriers, resistance of educators to change curricula, lack of staff with teaching experience in AI, and limited digitalization are some of the main barriers for implementing AI into the medical curriculum [3,10,12,13].”

In the citing paper

“Seven studies met the ROBIS criteria for low risk [12,13,38,39,40,41,42], one study was of unclear risk [43], and three studies were of high risk [44–46].”

In the citing paper

“Overall, this includes learning the AI terminology [13,37,45], strengths and limitations of AI [12,13,16,37,42], risks of AI [12,13,17,37,39,42], controlling for bias [42], critical appraisal AI literature and tools [12,13,34,37,39,45,46], and shared decision-making [12,13,37,41,45].”

In the citing paper

“In particular, knowledge related to data stewardship [12,13,37,41], data preprocessing and acquisition [13], standardization [13], data analytics [13,17,39], health data infrastructure [45], programming [39,41], and big data [13,45], may help physicians appreciate the obstacles faced by data scientists and improve the multidisciplinary collaboration [13,39].”

In the citing paper

“Eleven studies suggest that physicians should understand the role of data science in AI [12,13,16,37,39,41,42,45].”

In the citing paper

“Quality improvement was also reflected by the need for evaluating and iterating on any AI curriculum designed for medical schools on a regular basis to further improve the learning experience [12].”

In the citing paper

“Quality improvement was reflected through the need for critical appraisal of AI literature and tools [12,13,34,37,39], physician involvement in AI tool design/development [13,39], the ability to identify and control for bias [45], communication with data scientists and engineers [13], and communication with patients [12,13,37,41,42].”

In the citing paper

“Furthermore, four key relationships for communication were identified: data scientist to physician [13], physician to physician/allied healthcare professional [12,13,42], physician to patient [12,13,37,41,42], and physician to payer [13,37].”

In the citing paper

“Other important skills include knowledge related to machine learning (ML) [12,13], deep learning (DL) [13], natural language processing (NLP) [13], model development [13], and AI tool design and development [13,39].”

In the citing paper

“The goal of this communication is to equip physicians with the tools needed to successfully use AI in a clinical setting [12,13,16,40,44,45].”

In the citing paper

“Three key types of communication were identified: communication for clinical decision-making [13], communication for implementation [13,37], and communication for knowledge dissemination [12,13,37,41,42].”

In the citing paper

“Communication for knowledge dissemination focuses on sharing knowledge to improve the understanding of AI tools with respect to their role in care, as well as the strengths, limitations, and risks of AI [12,13,17,37,39,42].”

In the citing paper

“This would equip them with the skills needed to critically appraise AI literature and tools before implementing them [12,13,34,37,39], interpret model output [13,37,42], and communicate AI findings to others [12,13,42].”

In the citing paper

“Studies suggest that ML, DL, and NLP should be included in the early years of training in undergraduate medical education in order to increase student exposure to AI [12,13].”

In the citing paper

“Lee et al. support this sentiment, suggesting the need for the development of a standardized list of core AI competencies and the implementation of an AI curriculum to medical schools [12].”

In the citing paper

“Nine studies discuss the need for a fundamental understanding of AI ethics to improve the adoption of AI in medicine and ensure its safe use [12,13,16,37,39,45].”

In the citing paper

“Furthermore, by providing medical students with an opportunity to learn about AI ethics before they enter their clinical experiences, it could not only better prepare them to work with AI, but also better communicate to patients about AI [12,40].”

In the citing paper

“At least 11 medical schools across the United States of America, Canada and the Republic of Korea have pilot AI curricula (3).”

In the citing paper

“Although other publications acknowledge the importance of ethics in AI education, they do not provide specific content or guidance [36-39].”