The Use of MRI-Derived Radiomic Models in Prostate Cancer Risk Stratification: A Critical Review of Contemporary...

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Radiomics, a quantitative analysis of medical images, is emerging in prostate cancer (PCa) research.
• Diagnostic magnetic resonance imaging (MRI) provides rich data for radiomic analysis.

Purpose of the Study:

• To review the literature on MRI-derived radiomics for PCa risk stratification.
• To assess the performance of radiomic models in predicting PCa outcomes.

Main Methods:

• A systematic literature search was conducted for studies published between 2017 and 2022.
• Identified 218 articles, focusing on 33 that developed PCa risk stratification models using MRI radiomics.

Main Results:

• Radiomic models demonstrated good to excellent prediction for Gleason score (ROC-AUC 0.50-0.92) and adverse pathology (ROC-AUC 0.60-0.92).
• Prediction of post-treatment recurrence and failure showed high accuracy (ROC-AUC 0.73-0.99).
• Only 21.2% of studies included external validation.

Conclusions:

• MRI-derived radiomics shows significant potential for PCa risk stratification, particularly for Gleason score and adverse pathology.
• Predicting treatment outcomes requires further investigation and validation.
• Clinical integration necessitates robust validation of radiomic models.