Using Motion-Tracking to Objectively Assess Skill in Surgery

Surgical assessment often relies on subjective expert observation, which may introduce bias. We aimed to objectively differentiate skill levels in a complex simulated operative procedure using hand-tracking metrics through a wearable mixed reality headset.

A validated paediatric open inguinal herniotomy simulator was used to simulate a surgical task composed of multiple sequential operative steps. Novices (medical students) and experts (paediatric surgeons) were recruited. After completing a pre-task questionnaire, participants performed the task while wearing the mixed reality headset (Microsoft HoloLens 2). A post-task questionnaire was completed. Metrics collected included total path length, average path length, average movement velocity and total time. Data are reported as mean (± SD) and analysis included descriptive statistics, frequency counts and comparative testing using t-tests. A p-value of < 0.05 was considered significant.

A total of 33 participants were recruited (9 experts, 24 novices). The expert group had a shorter mean total path length: 287.0 ± 60.0 cm versus 450.1 ± 55.7 cm (p < 0.0001). They also had a shorter mean path length; 71.8 ± 15.1 cm versus 112.5 ± 13.9 cm (p < 0.0001). The expert mean velocity of movement was 0.17 ± 0.07 cm/s versus 0.14 ± 0.02 cm/s in novices (p = 0.13). Experts took less time to complete the task: 458.1 ± 65.8 s versus 836.0 ± 70.0 s (p < 0.0001).

This study objectively demonstrated that experts had a more efficient economy of movement and that hand-tracking metrics can differentiate between novice and expert performance during a simulated complex surgical task.