Artificial intelligence for assessment in competency-based medical education: current practices and future directions
Postgraduate Medical Journal · 21.08.2026 · 102 (1211) · 806-816
DOI: 10.1093/postmj/qgaf195 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Postgraduate Medical Journal, online erschienen am 18.05.2026. Verfasst von Lucy Muying Hui und 3 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Medicine, University of British Columbia, 2194 Health Sciences Mall, Vancouver, British Columbia, V6T 1Z3, Canada. Hauptthemen laut MEDLINE: Competency-Based Education, Artificial Intelligence, Clinical Competence, Educational Measurement. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 39 Literaturangaben.
Abstract (Originalsprache)
Background
Competency-Based Medical Education (CBME) relies on frequent, competency-focused assessments, which can be challenging to implement consistently. Artificial Intelligence (AI) holds promise to improve assessment efficiency, objectivity, and feedback in CBME, but its use remains in early stages with limited understanding of current practices and evaluation methods. This study aims to map existing AI applications in CBME assessments to guide future work.
Methods
A comprehensive search was performed in MEDLINE (Ovid), EMBASE (Ovid), PsycINFO, and Scopus using tailored keywords and MeSH terms. Included studies focused on the deployment of AI for assessment within CBME, covering applications in generating, analyzing, or interpreting evaluation data across undergraduate, graduate, and continuing professional education. The PRISMA-ScR guidelines were used to ensure transparent reporting, and findings were synthesized following Levac et al.’s approach.
Results
Of the 1002 search results, 32 studies met the inclusion criteria. Key findings indicate a wide application of AI from surgical or procedural skill assessment, to clinical note assessment, communication assessment, feedback generation, projected trainee performance, and analysis of narrative feedback from supervisors.
Conclusion
This review highlights potential advantages, such as timely evaluations, and challenges, such as lack of granularity, of AI integration. In conclusion, thoughtful integration of AI into competency-based medical education can complement traditional assessment methods and enhance learner outcomes, provided it is supported by robust infrastructure, ethical oversight, and collaborative policy development.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Faculty of Medicine, University of British Columbia, 2194 Health Sciences Mall, Vancouver, British Columbia, V6T 1Z3, Canada.
- Faculty of Medicine, Queen's University, 751 King Street West, Kingston, Ontario, K7L 3J7, Canada.
ORCID-Profile: Lucy Muying Hui
Zitieren
APA
Hui, L. M., Yu, E., Chung, A., & Kwan, B. Y. M. (2026). Artificial intelligence for assessment in competency-based medical education: current practices and future directions. Postgraduate Medical Journal, 102(1211), 806-816. https://doi.org/10.1093/postmj/qgaf195
Vancouver
Hui LM, Yu E, Chung A, Kwan BYM. Artificial intelligence for assessment in competency-based medical education: current practices and future directions. Postgraduate Medical Journal. 2026;102(1211):806-816. doi:10.1093/postmj/qgaf195
BibTeX
@article{Hui2026,
author = {Hui, Lucy Muying and Yu, Enoch and Chung, Andrew and Kwan, Benjamin Y M},
title = {Artificial intelligence for assessment in competency-based medical education: current practices and future directions},
journal = {Postgraduate Medical Journal},
year = {2026},
volume = {102},
number = {1211},
pages = {806--816},
doi = {10.1093/postmj/qgaf195}
}
EUVIMED