Related Experiment Video
Updated: Jun 23, 2025
Robotic Left Hepatectomy using Indocyanine Green Fluorescence Imaging for an Intrahepatic Complex Biliary Cyst
Published on: June 24, 2022
Impact of an AI-based laparoscopic cholecystectomy coaching program on the surgical performance: a randomized
Shangdi Wu1,2, Ming Tang1,2, Jie Liu3
1Department of General Surgery, Division of Pancreatic Surgery, West China Hospital of Sichuan University.
Background:
Laparoscopic cholecystectomy (LC) is the gold standard for treating symptomatic gallstones but carries inherent risks like bile duct injury. While the critical view of safety (CVS) is advocated to mitigate bile duct injury, its real-world adoption is limited. Additionally, significant variations in surgeon performance impede procedural standardization, highlighting the need for a feasible, innovative, and effective training approach. The aim of this study is to develop an artificial intelligence (AI)-assisted coaching program for LC to enhance surgical education and improve surgeon's performance.
Materials And Methods:
The authors conducted a multicenter, randomized controlled trial from May 2022 to August 2023 to assess the impact of an AI-based coaching program, surgical coaching program, on novice performing LC. Surgeons and patients meeting specific inclusion criteria were randomly assigned to either a coaching group with AI-enhanced feedback or a self-learning group. The primary outcome was assessed using the Laparoscopic Cholecystectomy Rating Form, with secondary outcomes including surgical safety, efficiency, and adverse events. Statistical analyses were performed using SPSS, with significance set at a P -value less than 0.05.
Results:
Between May 2022 and August 2023, 22 surgeons were initially enrolled from 10 hospitals, with 18 completing the study. No demographic differences were noted between coaching and self-learning groups. In terms of surgical performance (Laparoscopic Cholecystectomy Rating Form scores), the coaching group showed significant improvement over time (31 to 40, P =0.008), outperforming the self-learning group by study end (40 vs 38, P =0.032). Significant improvements in CVS achievement were also noted in the coaching group (11% to 78%, P =0.021). Overall, the coaching program was well-received, outpacing traditional educational methods in both understanding and execution of CVS and participants in the intervention group expressed strong satisfaction with the program.
Conclusions:
The AI-assisted surgical coaching program effectively improved surgical performance and safety for novice surgeons in LC procedures. The model holds significant promise for advancing surgical education.
