Computer vision tracking is the most common method of automated operative video analysis, with CNN the leading AI-based approach¶

Claim · All claims
Evidence · 1 study · review r3 · q2
Currency · newest study 2024 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Computer vision tracking was used in 7 of 13 studies, all open-surgery studies plus one endoscopic study, three of which required manual identification of the region of interest before automated tracking. → Dick L 2024 q2 i? Convolutional neural networking was the most common AI-based method (4 studies), with PSPN heat maps and a temporal shifting module also used. → Dick L 2024

Evidence¶

Dick L 2024¶

Dick L, Boyle CP, Skipworth RJE, Smink DS, Tallentire VR, Yule S. (2024). Automated analysis of operative video in surgical training: scoping review. BJS Open. https://doi.org/10.1093/bjsopen/zrae124

q2 · i? · review · r3

Descriptive synthesis of methods across the 13 included studies. The review reports computer vision tracking in 7 studies and, separately, that "convolutional neural networking (CNN) being the most common method (4 studies)" among AI-based approaches.

"The most common method of automated video analysis was computer vision tracking (7 studies). This was used in all of the studies that assessed open surgery and one of the studies that assessed endoscopic procedures"

Discussion¶

Related Claims¶

• Movement and proximity metrics relate to team situation awareness: teams with higher rated SA showed less movement, and time spent near the primary surgeon predicted NOTSS scores — related
• Computer vision and machine learning enable objective, automated surgical skill assessment — related