Key result
Marker-less hand tracking using a generalized additive model predicted expert ratings of surgical fluidity of motion with an R^2 of 0.77.
Why the study?
Surgical performance assessments are costly and time consuming, motivating automatic quantification of hand motion to reduce the burden of surgical coaching.
Can marker-less hand tracking and machine learning models predict expert-rated surgical performance in benchtop suturing and tying tasks?
Observational (n=219)
Can marker-less hand tracking and machine learning models predict expert-rated surgical performance in benchtop suturing and tying tasks?
Effect estimate: R^2 0.77
Marker-less hand tracking combined with generalized additive models can accurately predict expert ratings of surgical motion fluidity and economy in benchtop tasks.
May enable objective skill assessment in surgical simulation; leaves open validation in clinical settings.
Objective: This study creates linear and generalized additive models (GAMs) of video-recorded two-dimensional hand motion (synonymously referred to as hand movements or hand kinematics) to predict expert-rated performance along a series of surgical motion scales. Background: Surgical performance assessments are costly and time consuming. Automatically quantifying hand motion may offload some burden of surgical coaching and intervention by automatically collecting features of psychomotor performance. Methods: Five experts rated anonymized video clips of benchtop suturing and tying tasks (n = 219) along four visual-analog (0-10) performance scales: fluidity of motion, motion economy, tissue handling, and hand coordination. Custom software tracked both participant hands across successive video frames and populated a robust feature set to train a series of predictive models to reproduce the expert ratings. Results: A GAM (which accounts for nonlinear effects) predicted fluidity of motion ratings with slope = 0.71, intercept = 1.98, and R 2 = 0.77 for clinicians of different experience levels. Fluidity of motion and motion economy models outperformed those created to predict hand coordination and tissue handling ratings. Conclusions: Hand motion tracking may not address all contextual features of surgical tasks. Future work will explore how well simulation-based models extrapolate to more dynamic settings of the operating room.
No takes yet. Share an insight, caveat, or question.
Azari et al. (2020) conducted an observational in Surgical performance (n=219). Marker-less hand tracking and predictive models vs. Expert ratings was evaluated on Prediction of expert-rated performance (fluidity of motion) (R^2 0.77). Marker-less hand tracking using a generalized additive model predicted expert ratings of surgical fluidity of motion with an R^2 of 0.77.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: