Interpretable Machine Learning Model Based on Intra- and Peritumoral MRI Radiomics for Predicting Biochemical Recurrence After Radical Prostatectomy

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Purpose To develop a predictive model for biochemical recurrence (BCR) after radical prostatectomy (RP) by integrating intratumoral and peritumoral MRI radiomics features with clinical independent risk factors. Methods This retrospective study analyzed 277 RP patients with complete follow-up data (≥1 year) from our institution, randomly divided into training (n=193) and test (n=84) cohorts. Regions of interest (ROIs) were manually delineated on T2-weighted imaging(T2WI) and apparent diffusion coefficient (ADC) maps. Peritumoral ROIs were expanded by 4 mm using Python and manually adjusted to exclude non-prostatic tissues. Radiomics models (Intra, Peri_4mm, IntraPeri_4mm), a clinical model (Clinic), and a combined radiomics-clinical model (Combined) were constructed. The predictive performance of these models was evaluated using different indexes. SHapley Additive exPlanations (SHAP) analysis was employed to visualize and interpret the decision-making process. Results Independent risk factors for BCR included extraprostatic extension (EPE), clinical N stage (N), seminal vesicle invasion (SVI), PI-RADS score, neutrophil-to-lymphocyte ratio (NLR), and maximum transverse diameter of prostate (MTD). The Clinic model achieved AUC of 0.897 (training) and 0.731 (test). The IntraPeri_4mm_ADC model showed AUC of 0.902 and 0.706, while the IntraPeri_4mm_T2 model yielded 0.842 and 0.662. The Combined model outperformed others (AUC: 0.978 and 0.810). DCA confirmed its higher net benefit. SHAP analysis revealed EPE as the top contributor to BCR prediction, followed by the ADC-derived radiomics score (ADC label_0). Conclusions The combined MRI radiomics-clinical model effectively predicts BCR post-RP. SHAP interpretability transforms "black-box" predictions into quantifiable feature contributions, aiding clinicians in risk stratification and personalized treatment planning.

Article activity feed