Copyright
©The Author(s) 2025. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastrointest Oncol. Dec 15, 2025; 17(12): 112873
Published online Dec 15, 2025. doi: 10.4251/wjgo.v17.i12.112873
Published online Dec 15, 2025. doi: 10.4251/wjgo.v17.i12.112873
Machine learning survival prediction in esophageal cancer using radiomics and body composition from pretreatment and follow-up T12-level computed tomography
Ming-Cheng Liu, Wen-Hsien Chen, Department of Medical Imaging, Taichung Veterans General Hospital, Taichung 407, Taiwan
Ming-Cheng Liu, Department of Medical Imaging and Radiological Sciences, Central Taiwan University of Science and Technology, Taichung 406, Taiwan
Yung-Yin Cheng, Department of Medical Imaging, Chung Shan Medical University Hospital, Taichung 402, Taiwan
Yung-Yin Cheng, Shao-Chieh Lin, Chun-Han Liao, Chia-Hong Hsieh, Mei-Fang Hsieh, Ph.D. Program of Electrical and Communications Engineering, Feng Chia University, Taichung 407, Taiwan
Chih-Hung Lin, Cheng-Yen Chuang, Division of Thoracic Surgery, Department of Surgery, Taichung Veterans General Hospital, Taichung 407, Taiwan
Wen-Hsien Chen, Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung 402, Taiwan
Chun-Han Liao, Division of Medical Imaging, Yuanlin Christian Hospital, Changhua 510, Taiwan
Chun-Han Liao, Mei-Fang Hsieh, Department of Medical Imaging, Changhua Christian Hospital, Changhua 500, Taiwan
Chia-Hong Hsieh, Department of Radiology, Taichung Armed Forces General Hospital, Taichung 411, Taiwan
Yi-Jui Liu, Department of Automatic Control Engineering, Feng Chia University, Taichung 407, Taiwan
Author contributions: Liu MC, Cheng YY, Lin CH, Chuang CY, Chen WH, Liao CH, Hsieh CH, and Hsieh MF provided expertise in esophageal cancer and sarcopenia; Lin SC and Liu YJ contributed to the radiomics analysis and development of the machine learning models; Liu MC and Liu YJ were responsible for study design and literature review; Liu MC, Cheng YY, Lin SC, and Lin CH collected and organized patient data; Liu MC, Lin SC, and Liu YJ drafted the manuscript. All authors reviewed, revised, and approved the final version of the manuscript.
Supported by Taiwan National Science and Technology Council, No. NSTC114-2221-E-035-036; and Taichung Veterans General Hospital/Feng Chia University Joint Research Program, No. TCVGH-FCU1148207.
Institutional review board statement: This study was designed as a retrospective review, approved as a completely ethical review by the Institutional Review Board Taichung Veterans General Hospital (Approval No. CE25594B).
Informed consent statement: Approval from the Institutional Review Board of Taichung Veterans General Hospital was obtained, and the requirement for informed consent was waived.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: No additional data are available.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Yi-Jui Liu, PhD, Professor, Department of Automatic Control Engineering, Feng Chia University, No. 100 Wenhwa Road, Seatwen, Taichung 407, Taiwan.
Received: August 8, 2025
Revised: September 28, 2025
Accepted: October 28, 2025
Published online: December 15, 2025
Processing time: 125 Days and 5.7 Hours
Revised: September 28, 2025
Accepted: October 28, 2025
Published online: December 15, 2025
Processing time: 125 Days and 5.7 Hours
Core Tip
Core Tip: This study introduces a novel prognostic approach using radiomics and body composition analysis features extracted at the T12 vertebral level from pretreatment and follow-up computed tomography scans in esophageal cancer patients. Unlike conventional methods relying on L3-level imaging, this model incorporates T12-based skeletal muscle and adipose metrics - readily available in standard chest computed tomography - combined with clinical data to predict overall survival. The combined model incorporating clinical, body composition analysis, and radiomic data achieved excellent prognostic accuracy (area under the time-dependent receiver operating characteristic curve = 0.91) in 2-year survival prediction. This method supports non-invasive, automated, and personalized risk stratification, especially when follow-up imaging lacks L3 coverage.