Technical skills assessment: The expert versus the algorithm

(link to: Technical skills assessment: The expert versus the algorithm - Dick - 2024 - The Clinical Teacher - Wiley Online Library

This article explores the growing use of artificial intelligence-based assessment (AIBA) in healthcare education, particularly for evaluating technical skills. While expert-based assessment (EBA) is commonly used, AIBA offers potential advantages, such as improved efficiency and cost-effectiveness. Studies have shown correlations between AIBA and EBA, suggesting AIBA's ability to assess technical proficiency, though it lacks the ability to provide detailed feedback. A hybrid approach combining both methods may be most effective. Despite limitations, such as reduced context awareness and challenges with validity and implementation, AIBA is likely to become more prevalent in technical skills assessment.

Identifying novel metrics of operative performance in surgical training: protocol for a Delphi consensus study

(link to: Identifying novel metrics of operative performance in surgical training: protocol for a Delphi consensus study | Journal of Surgical Protocols and Research Methodologies | Oxford Academic (oup.com)

This study aims to identify the most relevant objective performance metrics for surgical trainees and determine how they should be applied in training. Traditional assessment methods are time-consuming and prone to bias, while digital technology offers more objective, scalable measures of operative performance. A three-round electronic Delphi study will gather input from trainees, trainers, researchers, and institutions to reach consensus on these metrics.

Automated analysis of operative video in surgical training: Scoping review

[In press]

This scoping review explores the use of automated video analytics in surgical training, identifying 13 studies that applied this technology to assess trainee-performed operations. Most studies used computer vision tracking to generate metrics on instrument movement and operation phases, with some evaluating critical safety elements. Automated metrics successfully distinguished skill levels and correlated with traditional assessments. However, there is a lack of research on the long-term impact of these tools on training outcomes, indicating a need for further investigation.