Radiomic-based machine learning model for the accurate prediction of prostate cancer risk stratification

Abstract

Objectives: To precisely predict PCa risk stratification, we constructed a machine learning (ML) model based on magnetic resonance imaging (MRI) radiomic features. Methods: Between August 2016 and May 2021, patients with histologically proven PCa who underwent preoperative MRI and prostate-specific antigen screening were included. The patients were grouped into different risk categories as defined by the EAU-EANM-ESTRO-ESUR-SIOG guidelines. Using Artificial Intelligence Kit software, PCa regions of interest were delineated and radiomic features were extracted. Subsequently, predictable models were built by utilizing five traditional ML approaches: support vector machine (SVM), logistic regression (LR), gradient boosting decision tree (GBDT), k-nearest neighbour (KNN) and random forest (RF) classifiers. The classification capacity of the developed models was assessed by area under the receiver operating characteristic curve (AUC) analysis. Results: A total of 213 patients were enrolled, including 16 low-risk, 65 intermediate-risk, and 132 high-risk PCa patients. The risk stratification of PCa could be revealed by MRI radiomic features, and second-order features accounted for most of the selected features. Among the five established ML models, the RF model showed the best overall predictive performance (AUC = 0.87). After further analysis of the subgroups based on the RF model, the prediction of the high-risk group was the best (AUC = 0.89). Conclusions: This study demonstrated that the MR radiomics-based ML method could be a promising tool for predicting PCa risk stratification precisely. Advances in knowledge: The ML models have valuable prospect for accurate PCa risk assessment, which might contribute to customize treatment and surveillance strategies.

Prostate Region-Wise Imaging Biomarker Profiles for Risk Stratification and Biochemical Recurrence Prediction

“…A similar approach with similar results associating Haralick features from the PZ with BCR was described by Gnep et al [ 19 ]. In terms of risk stratification, the utility of radiomics has been demonstrated in several studies in which machine learning models were able to accurately assess PCa risk [ 42 , 43 , 44 ].…”

Research on texture images and radiomics in urology: a review of urological MR imaging applications

“…MRI-radiomics have also been investigated in CaP risk stratification, to help customize treatment strategies, active surveillance vs curative. Indeed, Shu et al assessed the role of 5 ML models based on MRI radiomic features for predicting d’Amico group-risks, in a cohort of 213 patients, and observed that the random forest algorithm had the best performance with AUCs for the 3 groups significantly >0.8 (AUC = 0.89 for the high-risk group) [23].…”