Comparing Behavioral and Neural Activity Changes During Laparoscopic and Robotic Surgery Trainings

摘要

• Cognitive workload levels of general surgery residents were compared by measuring prefrontal cortex hemodynamic activity while performing a similar task using robot-assisted surgery RAS and laparoscopic surgery simulators. • Measures of interest included time to transfer each peg, change in oxygenated hemoglobin over the prefrontal cortex, and neural indices of relative neural involvement (RNI) and relative neural engagement (RNE) in each training simulation. • In addition to performance metrics, incorporating neural indices may provide insight into a trainee's experience during simulation training. This study aims to compare the cognitive workload levels of general surgery residents by measuring prefrontal cortex hemodynamic activity while performing a similar task using robotic-assisted surgery (RAS) and laparoscopic surgery simulators. The study was conducted with 22 general surgery residents who completed a peg transfer task in simulated laparoscopic and RAS training environments. Participants' performance and neurophysiological data were collected over a 1-month period. The study was conducted at Acibadem Mehmet Ali Aydinlar University- CASE (Center of Advanced Simulation and Education), utilizing a laparoscopic training simulator (CAE Lap VR) and robotic surgery simulator (Da Vinci Surgical System Si console with Backpack). Twenty-two general surgery residents (mean [SD] age, 29.45 [2.40] years; 18 [81.82%] male) volunteered for the study. None of the participants had prior experience with RAS or RAS simulators, whereas most had varying degrees of laparoscopic surgery experience. Significant differences were observed between RAS and laparoscopic simulations in terms of performance time and neural activity. Peg transfer times were shorter in RAS simulations compared to laparoscopic simulations (χ2(3) = 134.805, p < 0.001). Mean oxygenated hemoglobin (ΔHbO) levels in the prefrontal cortex were lower in RAS simulations (χ2(3) = 20.695, p < 0.001), indicating reduced cognitive workload. Relative Neural Efficiency (RNE) and Relative Neural Involvement (RNI) scores were higher in RAS tasks (χ2(1) = 55.765, p < 0.001), suggesting greater efficiency and involvement during robotic-assisted procedures. The findings indicate that RAS tasks are associated with lower cognitive workload and improved efficiency compared to laparoscopic tasks. Incorporating neural indices alongside performance metrics may enhance training assessments and provide deeper insights into trainees’ experiences in simulation-based surgical education.