©Author(s) (or their employer(s)) 2026. No commercial re-use. See Permissions. Published by Baishideng Publishing Group Inc.
World J Gastrointest Surg. Feb 27, 2026; 18(2): 113021
Published online Feb 27, 2026. doi: 10.4240/wjgs.v18.i2.113021
Published online Feb 27, 2026. doi: 10.4240/wjgs.v18.i2.113021
Prediction of lymphovascular invasion in rectal cancer based on multimodal magnetic resonance imaging radiomics model
Zheng-Hong Zhu, Yi Liang, Min Shi, Department of Radiology, General Hospital of the Yangtze River Shipping, Wuhan 430010, Hubei Province, China
Author contributions: Zhu ZH and Shi M designed the study; Zhu ZH and Liang Y collected and analyzed the clinical and imaging data; Zhu ZH performed radiomics feature extraction and model construction, wrote the original manuscript; Liang Y contributed to statistical analysis; Shi M supervised the project and critically revised the manuscript; all authors have read and approved the final manuscript.
Supported by Key Science and Technology Project of Changjiang River Administration of Navigational Affairs in 2025, No. 2025-CHKJ-014.
Institutional review board statement: This study was approved by the Institutional Review Board of General Hospital of the Yangtze River Shipping, Wuhan Brain Hospital, No. YL2024007.
Informed consent statement: The requirement for individual informed consent was waived by the Institutional Review Board because this retrospective study involved minimal risk, used de-identified data, and did not affect patient care.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: De-identified clinical data, the radiomics feature set, and analysis code are available from the corresponding author upon reasonable request. Raw magnetic resonance imaging images cannot be shared publicly due to patient privacy and institutional policies.
Corresponding author: Min Shi, MD, Department of Radiology, General Hospital of the Yangtze River Shipping, No. 5 Huiji Road, Jiang’an District, Wuhan 430010, Hubei Province, China. 15307154825@163.com
Received: October 10, 2025
Revised: November 5, 2025
Accepted: December 24, 2025
Published online: February 27, 2026
Processing time: 138 Days and 22.7 Hours
Revised: November 5, 2025
Accepted: December 24, 2025
Published online: February 27, 2026
Processing time: 138 Days and 22.7 Hours
Core Tip
Core Tip: This study developed a machine learning model integrating multimodal magnetic resonance imaging radiomics features and clinical factors to predict lymphovascular invasion in rectal cancer before surgery. Using fat-suppressed T2-weighted imaging, diffusion-weighted imaging, and contrast-enhanced T1-weighted imaging, feature-level fusion achieved high predictive performance. The combined clinical-radiomics model demonstrated the best accuracy (area under the curve = 0.867), offering a noninvasive, quantitative tool to guide individualized treatment planning.