Artificial intelligence in gastrointestinal and hepatopancreatobiliary surgery: innovations, integration, and future directions

Artificial intelligence (AI) is rapidly emerging as a transformative force in gastrointestinal (GI) and hepatopancreatobiliary (HPB) surgery, with the potential to enhance decision-making and surgical precision across the entire perioperative continuum. This narrative review explores the current applications and future directions of AI in GI and HPB surgical practice. We evaluated validated AI tools and platforms used in diagnostics, preoperative planning, intraoperative guidance, postoperative monitoring, and surgical education, while also considering the associated ethical and infrastructural challenges. AI-driven risk stratification models such as MySurgeryRisk and the POTTER calculator are improving preoperative prediction of complications. In endoscopy, systems like GI Genius have demonstrated improved adenoma detection rates, while EndoBRAIN-Plus enables real-time histologic assessment using high-resolution endocytoscopy. For surgical planning, virtual hepatectomy and three-dimensional reconstruction platforms such as HepaVision facilitate personalized and function-oriented liver resections. Intraoperatively, augmented reality and computer vision technologies provide enhanced anatomical visualization and guidance during minimally invasive procedures. AI-based predictive models are also contributing to improved postoperative monitoring by identifying patients at risk of complications such as anastomotic leaks. Furthermore, AI-driven simulation platforms are increasingly being integrated into surgical training and education. Overall, AI should be viewed not as a replacement for surgeons but as a powerful tool that can augment clinical judgment and improve surgical safety and personalization. However, challenges related to data standardization, model transparency, ethical governance, and clinical integration remain important barriers to widespread adoption.