Integration of multi-scale radiomics and deep learning for Ki-67 prediction in clear cell renal carcinoma.
Authors
Affiliations (9)
Affiliations (9)
- Department of Urology, First Hospital of Shanxi Medical University, No. 85 Jiefang South Road, Yingze District, Taiyuan, Shanxi, 030001, China.
- First Clinical Medical College of Shanxi Medical University, No. 56 Xinjian South Road, Yingze District, Taiyuan, Shanxi, 030001, China.
- Department of Medical Imaging, First Hospital of Shanxi Medical University, No. 85 Jiefang South Road, Yingze District, Taiyuan, Shanxi, 030001, China.
- College of Medical Imaging, Shanxi Medical University, No. 56 Xinjian South Road, Yingze District, Taiyuan, Shanxi, 030001, China.
- Department of Urology, Shanxi Provincial People's Hospital, No. 82 Shuangta Temple Street, Yingze District, Taiyuan, Shanxi, 030001, China.
- Department of Radiology, Shanxi Bethune Hospital, No. 99 Longcheng Avenue, Xiaodian District, Taiyuan, Shanxi, 030032, China.
- Department of Medical Imaging, First Hospital of Shanxi Medical University, No. 85 Jiefang South Road, Yingze District, Taiyuan, Shanxi, 030001, China. [email protected].
- Department of Urology, First Hospital of Shanxi Medical University, No. 85 Jiefang South Road, Yingze District, Taiyuan, Shanxi, 030001, China. [email protected].
- First Clinical Medical College of Shanxi Medical University, No. 56 Xinjian South Road, Yingze District, Taiyuan, Shanxi, 030001, China. [email protected].
Abstract
High Ki-67 expression in clear cell renal cell carcinoma (ccRCC) predicts poor prognosis but requires postoperative assessment. In a multicenter retrospective study of 627 ccRCC patients, we developed and validated a multi-modal model, integrating multi-scale radiomics and deep learning (DL) features, for non-invasive, preoperative Ki-67 prediction. Using ensemble machine learning algorithms, unimodal models were constructed from preoperative CT-derived multi-scale radiomics (intratumoral, habitat, peritumoral), 2D/3D DL, and clinical features. A stacking strategy was used to fuse the best-performing unimodal models. The fusion model demonstrated superior performance, achieving an Area Under the Curve (AUC) of 0.756 (95% CI 0.692-0.821) in the external test set. The model demonstrated excellent calibration and the highest clinical net benefit, with habitat radiomics identified as the dominant predictive component via SHAP analysis. Our validated multi-modal model significantly improves the preoperative prediction of Ki-67 expression compared to unimodal approaches, offering a promising tool to guide individualized surgical and surveillance strategies.