Skip to content
Review Open access

Artificial Intelligence for Colorectal Cancer Diagnosis: A Comprehensive Survey of Modalities, Frameworks, and Clinical Tasks

Nov 2026 · Cancers · Vol 18 · 0 citations · 169 references
Medicine

Abstract

Simple Summary Colorectal cancer is a leading cause of cancer-related deaths globally, making early and accurate diagnosis essential. The diagnostic journey involves multiple modalities, including stool tests, cross-sectional imaging, endoscopy, and tissue pathology. Artificial intelligence (AI) is increasingly being integrated into these steps to improve diagnostic accuracy and clinical efficiency. This comprehensive review synthesizes findings from 160 studies to evaluate the current state of AI across nine diagnostic modalities and seven specific clinical tasks. Our analysis reveals that AI tools for real-time polyp detection and optical diagnosis during colonoscopies are the most clinically mature, supported by strong evidence from randomized controlled trials. In contrast, emerging applications such as predicting tumor genetics from pathology slides or forecasting metastasis from imaging remain in the retrospective validation phase. We also outline a conceptual multimodal AI pipeline and identify key barriers to clinical adoption, including domain shift, data scarcity, and the need for standardized reporting frameworks. That said, this review highlights that while AI holds immense potential to transform colorectal cancer diagnostics, robust external validation, uncertainty quantification, explainability, and seamless integration into clinical workflows are critical prerequisites for routine patient care.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.