Global trends in artificial intelligence applications for gastric cancer prediction, treatment, and management: a topic modeling and bibliometric analysis.

PURPOSE: Artificial intelligence (AI) emerged as a promising tool for enhancing healthcare delivery and outcomes for gastric cancer (GC) patients. This study aimed to analyze research development, patterns, and trends in AI applications on GC. METHODS: In total, 1,854 publications from the Web of Science database (1993-2024) were extracted. Text-mining and bibliometric tools were employed to examine co-occurrence networks, global collaborations, and frequently used terms. Latent Dirichlet Allocation and dendrogram analysis identified hidden topics and clustered key research domains, while linear regression models evaluated research trends. RESULTS: This study highlighted a growing interest in AI applications for GC, with China, Japan, and the United States as leading contributors. However, an unequal distribution of research interest is disadvantaging low- and middle-income countries (LMICs) despite their higher cancer burden. Topic modeling revealed that Comparative Analysis of Robotic vs. Laparoscopic Gastrectomy Topic is the most studied area, though its interest has declined over the past five years. Emerging areas include AI applications in Tumor Segmentation, Predictive Models, Dietary and Environmental GC Prevention, Cell-Free DNA and Biomarkers. This study emphasized the AI's interdisciplinary nature, along with integrating computer science with clinical fields to advance GC diagnosis and treatment. CONCLUSION: The findings suggested that policymakers should prioritize targeted funding and equitable AI adoption in LMICs to reduce disparities. Furthermore, strategic efforts should be directed toward improving the quality of life for GC patients. Future studies should focus on addressing AI algorithm's limitations and building effective AI tools to improve treatment outcomes for population-specific demands.