Interpretable habitat and peritumoral radiomics from multiparametric MRI for preoperative high-risk prostate cancer prediction: a multi-institutional study.
Authors
Affiliations (10)
Affiliations (10)
- Department of Radiology, Jiangdu People's Hospital Affiliated to Yangzhou University, 100 Jiangzhou Road, Jiangdu district, Yangzhou, Jiangsu, 225200, PR China. [email protected].
- Medical College of Yangzhou University, 136 Jiangyang Middle Road, Hanjiang district, Yangzhou, Jiangsu, 225009, China. [email protected].
- Department of Radiology, Jiangdu People's Hospital of Yangzhou, 100 Jiangzhou Road, Jiangdu district, Yangzhou, Jiangsu, 225200, China. [email protected].
- Department of Radiology, Zhongda Hospital, Cultivation and Construction, State Key Laboratory of Intelligent Imaging and Interventional Medicine, Medical School of Southeast University, 87 Dingjiaqiao Road, Nanjing, Jiangsu, 210009, China.
- Department of Neurology, Jiangdu People's Hospital Affiliated to Yangzhou University, 100 Jiangzhou Road, Jiangdu district, Yangzhou, Jiangsu, 225200, China.
- Department of Cardiology, Jiangdu People's Hospital Affiliated to Yangzhou University, 100 Jiangzhou Road, Jiangdu district, Yangzhou, Jiangsu, 225200, China.
- Department of Radiology, Affiliated Hospital of Yangzhou University, Hanjiang Middle Road 368, Hanjiang district, Yangzhou, 225000, PR China.
- Department of Radiology, Jiangdu People's Hospital Affiliated to Yangzhou University, 100 Jiangzhou Road, Jiangdu district, Yangzhou, Jiangsu, 225200, PR China.
- Department of Orthopaedic Surgery, Yangzhou Hongquan Hospital, Huangshan Road 366, Jiangdu district, Yangzhou, 225000, PR China.
- Department of Pathology, Jiangdu People's Hospital Affiliated to Yangzhou University, 100 Jiangzhou Road, Jiangdu district, Yangzhou, Jiangsu, 225200, China.
Abstract
Current preoperative assessment faces limitations, including PI-RADS scoring subjectivity and diagnostic uncertainty in distinguishing high-risk prostate cancer from benign and low-risk lesions. To develop an interpretable ensemble learning framework integrating habitat-based radiomics and peritumoral analysis from multiparametric MRI for preoperative high-risk prostate cancer prediction. This retrospective, multi-institutional study included 896 patients with suspected prostate lesions and histopathologically confirmed diagnoses across three centers (January 2018-December 2024). Intratumoral habitat analysis used K-means clustering; peritumoral analysis evaluated 1 mm, 3 mm, and 5 mm expansion rings. Feature selection used minimum Redundancy Maximum Relevance (mRMR) and LASSO regression. Models were validated externally with SHAP analysis for interpretability. The cohort comprised 398 training, 171 internal validation, and 327 external validation patients. The habitat signature achieved superior performance with AUCs of 0.827 (95% CI: 0.768-0.886) and 0.855 (95% CI: 0.795-0.915) in external validation cohorts, significantly outperforming intratumoral signatures (AUCs: 0.774 and 0.629, p < 0.001) and clinical signatures (AUCs: 0.791 and 0.712, p < 0.001). The 3 mm peritumoral signature performed best (AUC: 0.782-0.793). The combined model achieved the highest performance (AUC: 0.860-0.876). SHAP analysis showed ADC-derived features dominated importance, with habitat region H3 contributing > 70% of selected features. Integrated habitat and peritumoral radiomics provide robust preoperative risk stratification for prostate cancer, with superior performance from ADC-derived habitat features. Not applicable. This was a retrospective observational study without prospective trial registration.