Accepted for/Published in: JMIR Medical Informatics
Date Submitted: Jun 6, 2025
Date Accepted: Oct 27, 2025
Predictive performance of radiomics-based machine learning for colorectal cancer recurrence risk: a meta-analysis
ABSTRACT
Background:
Predicting Colorectal Cancer (CRC) recurrence risk remains a challenge in clinical practice. Owing to the widespread use of radiomics in CRC diagnosis and treatment, some researchers recently explored the effectiveness of radiomics-based models in forecasting CRC recurrence risk. Nonetheless, the lack of systematic evidence of the efficacy of such models has hampered their clinical adoption.
Objective:
we explored the value of radiomics in predicting CRC recurrence, providing a scholarly rationale for developing more specific interventions.
Methods:
Four databases (Embase, PubMed, the Cochrane Library, and Web of Science) for relevant articles from inception to 1 January 2025. The quality of the included original studies was assessed using the Radiomics Quality Score (RQS). During the meta-analysis, subgroup analyses were conducted based on the validation and training sets.
Results:
This meta-analysis included 17 original studies involving 4600 CRC patients. In the validation set, the c-index values based on clinical features, radiomics features, and radiomics features+clinical features were 0.73 (95%CI:0.677-0.789), 0.80 (95%CI:0.746-0.846), and 0.83(95%CI:0.793-0.867), respectively. In the internal validation set, the c-index values based on clinical features, radiomics features, and radiomics features+clinical features were 0.70 (95%CI:0.612-0.791), 0.828 (95%CI:0.776-0.881), and 0.829 (95%CI:0.779-0.880), respectively. Finally, in the external validation set, the c-index values based on clinical features, radiomics features, and radiomics features+clinical features were 0.763 (95%CI:0.698-0.827), 0.747 (95%CI:0.659-0.834), and 0.831 (95%CI:0.778-0.884), respectively.
Conclusions:
Radiomics-based Machine Learning (ML) models, especially those integrating radiomics and clinical features, demonstrated notable clinical utility in predicting CRC recurrence risk. In addition to their predictive value, such integrated models could also be useful in developing specific intervention programs. Clinical Trial: Not applicable
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.