Related Experiment Video
Updated: Jan 13, 2026
Robotic Left Hepatectomy using Indocyanine Green Fluorescence Imaging for an Intrahepatic Complex Biliary Cyst
Published on: June 24, 2022
Quantifying surgical complexity: how smoothness metrics vary in robotic cholecystectomy
Elizabeth W Tindal1,2,3, Mattia Ballo4,5, Jeffrey E Nussbaum6,5
1Department of Surgery, Lenox Hill Hospital, Northwell Health, 186 E 76Th Street, 1St Fl, New York, NY, 10021, USA. liztindal@gmail.com.
Background:
Objective performance indicators (OPIs) include measures of instrument movement during robotic procedures and can assess movement smoothness. This study analyzes smoothness metrics in cholecystectomy, hypothesizing that longer, more complex tasks exhibit reduced movement smoothness.
Methods:
Cholecystectomy videos from our institution were analyzed, with Parkland grading used to assess cholecystitis severity. Since longer duration likely indicates greater complexity, cases were categorized as either complex or standard based on when the duration of a key operative step exceeded the surgeon's mean by at least one standard deviation - complex dissection (CD) vs standard dissection (SD). OPIs were evaluated for Calot triangle dissection (CTD) and liver bed dissection (LBD). Smoothness-related OPIs-including movement arrest period ratio (MAPR), spectral arc length (spalength), normalized speed (normspeed), speed peaks rate (SPR), and nondimensional jerk (nlognJerk)-were further classified based on arm function (e.g., dominant arm (DA) vs. active retraction (AR)).
Results:
Fifty-four videos were evaluated. Average step duration for CTD and LBD were 14.18 ± 9.87 and 10.05 ± 7.72 min, respectively. Case duration was correlated with Parkland grade (r = 0.26, p = 0.03). Parkland grade for CD was significantly higher than SD in CTD (2.7 vs 2.0, p < 0.03) but not in LBD. In both CTD and LBD, the smoothness profile for DA was affected differently than AR in CD. DA during CTD showed significant reductions in MAPR (0.78 vs 0.92), normspeed (0.24 vs 0.35), SPR (40.22 vs 65.63), nlognjerk (-37.96 vs -36.48) and spalength (-70,604.4 vs -35,742.5). However, significant changes were only seen for AR during CTD in SPR (11.73 vs 18.37), nlognjerk (-36.94 vs -35.23) and spalength (-119,126.5 vs -64,669.5) (all p < 0.05). Similar patterns were seen during LBD.
Conclusions:
Smoothness metrics seem to be affected by case difficulty and may be a marker for struggling and potentially used to determine when a surgeon needs help or guide case review.
