Artificial Intelligence Methods for the Differential Diagnosis of Irritable Bowel Syndrome and Inflammatory Bowel Disease: A Systematic Review

Abstract

Background and Aims: Differential diagnosis between irritable bowel syndrome (IBS) and inflammatory bowel disease (IBD) represents a major challenge in modern gastroenterology due to overlapping symptoms, limitations of traditional diagnostic methods, and the complexity of their pathophysiology. This review examines the application of artificial intelligence (AI) and machine learning (ML) methods to improve accuracy and efficiency in the differential diagnosis between IBS and IBD.
Methods: The review encompasses seven recent studies employing various AI/ML techniques, utilizing clinical, genetic, microbiomic, and imaging data.
Results: AI-based models exhibit high sensitivity and specificity, with remarkable performance by algorithms such as logistic regression, random forest, neural networks, and support vector machines. Highlighted biomarkers include long non-coding RNA molecules, DNA methylation profiles, and diverse compounds from gut microbiota.
Conclusions: Although AI/ML methods show significant potential for distinguishing IBS from IBD, existing studies present limitations, including small sample sizes, data heterogeneity, and generalizability challenges. The development of standardized protocols and extensive multicenter studies is recommended to clinically validate these models, facilitating their integration into current medical practice.