AI for colon cancer: A focus on classification, detection, and predictive modeling.

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Artificial Intelligence (AI) shows promise in enhancing colon cancer detection, classification, prediction, and segmentation.
• Study reliability hinges on the quality and completeness of research data.

Purpose of the Study:

• To systematically review AI applications in colon cancer research.
• To evaluate AI's impact on diagnostic accuracy, treatment planning, and patient outcomes.

Main Methods:

• Comprehensive literature search (PubMed, Scopus, Web of Science) from 2020-2024.
• Quality assessment of studies and meta-analysis where applicable.
• Subgroup analysis by AI type (deep learning, machine learning) and application; recording of Explainable AI (XAI) and Generative AI use.

Main Results:

• 80 articles reviewed, demonstrating AI's significant improvement in diagnostic accuracy for polyp detection and histopathology.
• Deep learning models generally outperformed traditional methods.
• Clinical integration faces challenges due to data and validation gaps.

Conclusions:

• AI offers substantial potential for advancing colon cancer diagnosis and treatment.
• Future efforts should prioritize clinical workflow integration of AI via explainable models and standardized validation.