Related Experiment Video
Updated: May 18, 2026
Use of Magnetic Resonance Imaging and Biopsy Data to Guide Sampling Procedures for Prostate Cancer Biobanking
Published on: October 10, 2019
Development and Validation of Radiomics and Clinical Models for Predicting Gleason Grades in Prostate Cancer Across
Li Liu1, Peiyuan Yin2, Mingbin Bao3
1Department of Radiology, Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, Chengdu 610072, Sichuan, China (L.L., B.H., F.L., L.Y.).
Rationale And Objectives:
To assess the value of radiomics models and clinical models (CM) based on diverse volumes of interest and clinical indicators in differentiating low Gleason grade group (GGG) from high-GGG in prostate cancer (PCa).
Materials And Methods:
This study included 312 PCa patients diagnosed pathologically from center 1 and center 2, divided into internal training (dataset A, n=144, center 1), internal validation (dataset B, n=63, center 1) and external test set (dataset C, n=105, center 2). The CM and radiomics models for intratumoral volume (ITV), 2 mm reduction from tumor border (CTV), 2 mm and 4 mm extensions beyond tumor border (PTV2, PTV4), and 4 mm peritumoral transition (PTT) were developed. Model performance was evaluated using area under the curve (AUC), net reclassification improvement (NRI) and integrated discrimination improvement (IDI).
Results:
The combined model ITV_PTV4_CM1 demonstrated optimal performance (AUC 0.871 [95% confidence interval: 0.802-0.938] for dataset C), significantly outperformed ITV (0.778 [0.690-0.864]), and PTV4 (0.774 [0.681-0.866]), with all p < 0.05, and also better than CM1 (0.830 [0.753-0.905]). The introduction of ITV+PTV4, PSAD+PSA, and PSA all provide positive gains to model (NRI/IDI > 0, all p < 0). PTT with AUCs of 0.858 and 0.869 for datasets A and B, outperformed other individual radiomics models in center 1. ADC/DWI_gldm emerged as the top-weighted feature.
Conclusion:
The ITV_PTV4_CM1 enhances predictive efficacy for preoperative PCa risk stratification. PTV4 and PTT highlight the importance of transition zone features at the peritumoral invasion margin, offering insights into optimal peritumoral extents. The similar efficacy of intratumoral and peritumoral radiomics suggests strategies for optimizing PCa diagnostic workflow.
