Application of Artificial Intelligence in Training and Assessment of Basic Endoscopic Surgical Skills

This article examines the application of artificial intelligence technologies for training and objectively assessing basic laparoscopic skills using the BÉSTA 2.0 simulation platform. The paper presents the system architecture, computer vision algorithms, stages of neural network training, and performance results. A three-phase study demonstrated high assessment accuracy (up to 95% agreement with expert ratings) and strong reproducibility of outcomes. Key limitations and mitigation strategies are discussed. Automated evaluation of the endosurgical skills enables scalable, instructor-independent feedback, improves training efficiency, and ensures real-time insight into trainee performance. The findings emphasize the potential of incorporating AI into laparoscopic training and, more broadly, into contemporary surgical education.