Current Role of Artificial Intelligence in the Management of Gastric Cancer

Abstract

Background/Objectives: In the era of precision medicine in gastric cancer, artificial intelligence has emerged as a tool in diagnosis, prognostic stratification, and clinical management. The role of big data analysis leveraging complex databases has been rapidly developing, thus challenging physicians to interpret and apply them in clinical practice. The aim of this comprehensive review is to present the current trends in the use of artificial intelligence in the field of diagnosis, histopathological evaluation, and clinical prognosis of gastric tumors. Methods: We screened the PubMed and MEDLINE databases for the latest studies with the development and evaluation of artificial intelligence algorithms in gastric cancer. Different sorts of deep learning protocols were explored, and their standardized applications are presented herein. Results: A broad spectrum of AI-based models extending from the surveillance of high-risk subepithelial lesions to the precise molecular characterization of tumors and treatment management are gaining space in clinical practice. However, all current studies are lacking a randomized design at the large-population scale, which is required to further integrate machine learning algorithms into standard clinical care. Conclusions: Despite the remaining challenges of data quality, algorithm improvement, and outcome interpretation, there is promising evidence that artificial intelligence can revolutionize gastric cancer management. There is a need to develop AI algorithms based on big data sources that must consequently be evaluated in randomized multicenter studies.