SRS370 - Identifying novel indicators of non-technical skills derived from operative video annotation

Abstract Background Non-technical skills (NTS), including situation awareness and decision-making, are essential for safe surgical care and are increasingly recognised as key contributors to patient outcomes. Current NTS assessments rely on expert observation, which although provides nuanced insights is limited by scalability and reproducibility. To address this, we investigated whether surgical gestures, derived from annotated video of the surgical field, could serve as objective indicators of cognitive NTS. Methods A dataset of 40 open-source laparoscopic appendicectomy videos was annotated for temporal (for example surgical gestures) and spatial (for example coordinates of actions) indicators of surgical NTS. Using the Non-Technical Skills for Surgeons (NOTSS) tool, twelve expert observers independently assessed situation awareness and decision-making. Multivariable linear regression identified video-derived indicators predictive of cognitive NTS. Results Across all videos, a total of 10 385 events were annotated, generating 87 374 data points. The mean number of events per video was 259.6 (SD 76.3) and the mean cognitive NOTSS rating was 5.6/8 (SD 0.8). The final multivariable model explained 39.6% of the variance in expert cognitive NOTSS ratings, identifying five predictors (ß estimate, 95% c.i.): dexterity index (−0.49, −0.8--0.18), events during the final operative phase (0.16, 0.04–0.28), coagulating events (−0.02,−0.03–0.01), dropping actions (0.08, 0.01–0.14) and actions targeting the small bowel (−0.01, −0.01–0.001). Conclusions This study provides the first evidence that surgical gestures can provide insights into higher-order thinking. These findings lay the foundations for scalable, automated tools to evaluate surgeons’ cognitive processes offering new avenues to improve surgical training, performance and outcomes.