Machine Learning for Colitis-Associated Cancer in Inflammatory Bowel Disease: Evidence and Future Directions Toward...

Area of Science:

• Gastroenterology and Oncology
• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Colitis-associated cancer (CAC) is a serious complication of inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn's disease (CD).
• The inflammation-dysplasia-carcinoma sequence in CAC differs from sporadic colorectal cancer, posing challenges for early detection and risk stratification.
• Current surveillance methods like endoscopy and histopathology are limited by time, operator dependency, and potential to miss early neoplastic changes.

Purpose of the Study:

• To review the current evidence on artificial intelligence (AI) and machine learning (ML) applications for CAC detection and risk prediction in IBD.
• To discuss the technical and clinical challenges associated with AI implementation in IBD surveillance.
• To highlight future directions for integrating AI into clinical practice for improved patient outcomes.

Main Methods:

• Review of existing literature on AI, ML, and deep learning (DL) applications in CAC detection and risk prediction within IBD.
• Analysis of studies evaluating AI's ability to identify dysplasia and CAC.
• Exploration of multimodal data integration (clinical, endoscopic, histological, molecular) for enhanced predictive performance.

Main Results:

• Early studies suggest AI/ML approaches can improve the detection of dysplasia and CAC in IBD patients.
• AI demonstrates potential in enhancing risk prediction models for CAC.
• Multimodal data integration with AI may lead to more accurate predictions and personalized medicine strategies.

Conclusions:

• AI offers a promising avenue to overcome limitations of conventional surveillance methods for CAC in IBD.
• Further research and validation are needed to integrate AI into routine clinical practice for improved IBD management.
• AI has the potential to significantly improve early detection, risk stratification, and clinical outcomes for IBD patients at risk of CAC.