497 Artificial Intelligence in Global Surgery: A Scoping Review of Primary Research from Low-and Middle-Income Countries

Aim Surgical care in low- and middle-income countries (LMICs) faces diagnostic and workforce constraints. Artificial intelligence (AI) could expand access to care but the scope of AI research and adoption for surgical care within LMICs is unclear. This scoping review synthesises primary studies to outline research activity and progress toward implementation. Method A prospectively registered scoping review following PRISMA-ScR guidelines (OSF.IO/9PV6A). PubMed, Scopus, and Web of Science were searched to 14/07/2025. We included primary research evaluating surgical applications of AI in LMICs, as defined by the World Bank. Data were extracted to describe geographical distribution and the maturity of AI systems using the OECD AI lifecycle framework. Results Of 2,602 records, 474 studies met inclusion criteria. Activity was highly concentrated with 64.6% of studies from China (n=306). Only 9.5% (n=45) originated from other LMICs and 0.8% (n=4) from low-income countries. Minimal evidence originated from Sub-Saharan Africa and Latin America. Most work focused on algorithm development and internal validation (n=317, 66.9%). External validation was reported in 111 studies (23.4%), while clinical deployment was reported in 12 (2.5%). There was no translation beyond the development stage in lower-income settings. Conclusions As the first scoping review on this topic, our findings highlight that primary research on AI in surgical care across LMICs remains limited and unevenly distributed, with marked gaps in low-income regions. Most studies originate from Asian higher-middle-income countries, with minimal clinical deployment. Without strengthened data infrastructure, technical capacity, and system readiness, AI may deepen rather than reduce global surgical inequities.