AI-powered intestinal ultrasound in inflammatory bowel disease: Advancing toward automated disease assessment

The management of inflammatory bowel disease (IBD) relies on accurate disease assessment and close monitoring to guide therapy and evaluate treatment response. While endoscopy remains the gold standard for detecting mucosal inflammation, its invasiveness and inability to fully assess transmural disease or extraintestinal complications limit its use as a stand-alone tool. Cross-sectional imaging modalities, such as CT and MRI, provide accurate evaluations but are hindered by high cost, limited accessibility, and patient burden. Therefore, intestinal ultrasound (IUS) has emerged as a safe, non-invasive, and reproducible modality to evaluate transmural inflammation and complications in real time, making it increasingly relevant for both Crohn's disease and ulcerative colitis. However, challenges such as operator dependence, variability in image acquisition, and lack of standardize protocols have limited its widespread adoption. Artificial intelligence (AI) offers transformative potential to overcome these barriers by automating image analysis, standardising interpretation, and reducing interobserver variability. Early applications have shown promise in bowel wall segmentation and thickness assessment, with future opportunities to further integrate imaging features with biomarkers, clinical data, and multi-omics for precision medicine. This review aims to explore the current and future roles of IUS, summarizing the potential of AI in IUS to improve diagnostic accuracy, enhance disease monitoring, and enable personalized patient management. We will discuss the strengths and limitations of existing approaches, highlight barriers to clinical implementation, and outline perspectives on how AI-enhanced IUS may transform both research and routine care in IBD.