An Online Teaching Resource to Support UK Medical Student Education During the COVID-19 Pandemic: A Descriptive Account

Karunaratne, Dilhara; Karunaratne, Nisal; Wilmot, Jade; Vincent, Tim; Wright, Juliet; Mahmood, Nadia; Tang, Alice; Sam, Amir H.; Reed, Malcolm; Howlett, David

doi:10.2147/amep.s337544

Cited by 3 publications

(1 citation statement)

References 10 publications

(16 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The covid-19 pandemic spurred a shift towards online learning, and more medical education than ever is being taught in a virtual format 5,6. This transition contributed to the accelerated growth of pre-existing online resources, including question banks such as UWorld and AMBOSS, as well as video based procedural content 7,8.…”

Section: Opportunities For Changementioning

confidence: 99%

Curriculum committees are a chance for medical students to drive educational change

Wahlstedt,

Danby

2024

BMJ

View full text Add to dashboard Cite

show abstract

Section: Opportunities For Changementioning

confidence: 99%

Curriculum committees are a chance for medical students to drive educational change

Wahlstedt,

Danby

2024

BMJ

View full text Add to dashboard Cite

show abstract

The Impact of Patient-Centric Interactive E-Module in Pathology Among Medical Undergraduates

Venkatesan,

Manickam,

Madasamy

et al. 2023

Med.Sci.Educ.

View full text Add to dashboard Cite

Which curriculum components do medical students find most helpful for evaluating AI outputs?

Waldock,

Lam,

Baptista

et al. 2024

Preprint

View full text Add to dashboard Cite

Introduction The risk and opportunity of Large Language Models (LLMs) in medical education both rest in their imitation of human communication. Future doctors working with generative artificial intelligence need to judge the value of any outputs from LLMs to safely direct the management of patients. We set out to evaluate our students’ ability to validate LLM responses to clinical vignettes, identify which prior learning they utilised to scrutinise the LLM answers, and whether they were aware of ‘clinical prompt engineering’. Methods A content analysis cohort study was conducted amongst 148 consenting final year medical students at Imperial College London. A survey asked students to evaluate answers provided by GPT 3.5 in response to ten clinical scenarios, five of which GPT 3.5 had answered incorrectly, and to identify which prior training enabled them to determine the accuracy of the GPT 3.5 output. Results The overall median student score in correctly judging the answers given by GPT 3.5 was 61%, with 65% demonstrating sound clinical reasoning for their decision. Students reported interactive case-based discussions and pathology teaching to be the most helpful for AI output evaluation. Only 5% were aware of ‘clinical prompt engineering’. Conclusion Artificial intelligence is a sociotechnical reality, and we need to validate the new pedagogical requirements for the next generation of doctors. Our data suggest that critical analysis taught by pathology clinical case teaching is currently the self-reported best training for medical students to evaluate the outputs of LLMs. This is significant for informing the design of medical training for future doctors graduating into AI-enhanced health services.

show abstract

An Online Teaching Resource to Support UK Medical Student Education During the COVID-19 Pandemic: A Descriptive Account

Cited by 3 publications

References 10 publications

Curriculum committees are a chance for medical students to drive educational change

Curriculum committees are a chance for medical students to drive educational change

The Impact of Patient-Centric Interactive E-Module in Pathology Among Medical Undergraduates

Which curriculum components do medical students find most helpful for evaluating AI outputs?

Contact Info

Product

Resources

About