Previously submitted to: JMIR Medical Informatics (no longer under consideration since May 19, 2026)
Date Submitted: Jul 30, 2025
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Development and Validation of a Dual-Path Intelligent Auxiliary Diagnostic System for Colorectal Cancer
ABSTRACT
Background:
Colorectal cancer (CRC) is a major global health concern, emphasizing the importance of early detection. This study aims to develop a dual-path AI strategy combining molecular analysis and image segmentation to enhance intelligent CRC diagnosis.
Objective:
To address existing screening limitations, including operator-dependency and missed detections, by integrating molecular and imaging approaches for improved CRC diagnosis.
Methods:
The study independently modeled two pathways: molecular analysis using transcriptomic data and machine learning models, and image segmentation employing the EMT-Net model with state-of-the-art metrics across various colonoscopy datasets.
Results:
Core diagnostic genes (e.g., CDC25B, TEAD4, MMP7) were identified with high stability and interpretability, achieving an AUROC of 0.987 and F1-score of 0.976. The EMT-Net model displayed superior segmentation performance compared to mainstream models on multiple datasets.
Conclusions:
The fusion of molecular and imaging pathways in the AI-based CRC diagnosis strategy offers methodological innovation and potential clinical applications. The molecular pathway enhances interpretability, while the imaging pathway enhances precision, collectively forming a foundation for advanced and generalizable early CRC screening platforms.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.