Role of Artificial Intelligence in Gastroenterology Training (2005–2025): Trends, Tools, and Challenges

Over the past two decades, gastroenterology training (GI training) has undergone a significant transformation through the integration of advanced technologies, particularly artificial intelligence (AI). The emergence of AI as a transformative tool has facilitated notable changes in how GI trainees acquire knowledge and skills. Key applications of AI in this context include simulation-based learning, diagnostic decision support, and procedural skill acquisition. These AI-driven innovations are increasingly recognised for enhancing learning efficiency, improving diagnostic accuracy, and building procedural confidence among trainees. To explore the extent of AI's impact on GI education, a comprehensive literature review was conducted. The search followed PRISMA guidelines and focused on peer-reviewed articles published between 2005 and 2025. Databases such as PubMed/NCBI, ScienceDirect, and the Cochrane Library were used, along with targeted searches in leading GI journals. The initial search yielded 312 records. After applying the inclusion and exclusion criteria, 22 studies were selected for final synthesis. These included randomised controlled trials, observational studies, systematic reviews, and narrative analyses. The reviewed studies consistently demonstrated that AI-enhanced simulation tools, particularly those incorporating virtual reality (VR) and augmented reality (AR), played a pivotal role in procedural training. These tools offered immersive, risk-free environments that allowed trainees to practice and refine their technical skills before applying them in real-world clinical scenarios. AI also proved valuable in diagnostic decision support. Systems such as computer-aided detection (CADe) were shown to significantly increase lesion detection rates during endoscopic procedures, contributing to improved clinical decision-making and better patient outcomes. Additionally, AI-assisted technologies enhanced procedural training by supporting more precise biopsy targeting and facilitating lesion identification during endoscopic ultrasound (EUS). As per our review, the evidence suggests that AI technologies are making meaningful contributions to GI training by improving diagnostic capabilities, streamlining the learning process, and supporting technical skill acquisition. However, despite these promising developments, further research is necessary. Future studies should include multi-centre randomised controlled trials and longitudinal evaluations to establish long-term efficacy. Furthermore, efforts toward global standardisation of AI training tools and equitable access are essential to ensure that these technologies benefit trainees across diverse clinical settings.