Artificial Intelligence in Gastrointestinal Surgery: A Systematic Review of Its Role in Laparoscopic and Robotic Surgery

Abstract

<b>Background</b>: Artificial intelligence (AI) is transforming surgical practice by enhancing training, intraoperative guidance, decision-making, and postoperative assessment. However, its specific role in laparoscopic and robotic general surgery remains to be clearly defined. The objective is to systematically review the current applications of AI in laparoscopic and robotic general surgery and categorize them by function and surgical context. <b>Methods</b>: A systematic search of PubMed and Web of Science was conducted up to 22 June 2025, using predefined search terms. Eligible studies focused on AI applications in laparoscopic or robotic general surgery, excluding urological, gynecological, and obstetric fields. Original articles in English or Spanish were included. Data extraction was performed independently by two reviewers and synthesized descriptively by thematic categories. <b>Results</b>: A total of 152 original studies were included. Most were conducted in laparoscopic settings (<i>n</i> = 125), while 19 focused on robotic surgery and 8 involved both. The majority were technical evaluations or retrospective observational studies. Seven thematic categories were identified: surgical decision support and outcome prediction; skill assessment and training; workflow recognition and intraoperative guidance; object or structure detection; augmented reality and navigation; image enhancement; technical assistance; and surgeon perception and preparedness. Most studies applied deep learning, for classification, prediction, recognition, and real-time guidance in laparoscopic cholecystectomies, colorectal and gastric surgeries. <b>Conclusions</b>: AI has been widely adopted in various domains of laparoscopic and robotic general surgery. While most studies remain in early developmental stages, the evidence suggests increasing maturity and integration into clinical workflows. Standardization of evaluation and reporting frameworks will be essential to translate these innovations into widespread practice.

Ähnliche Arbeiten