Interpretable Prostate Cancer Detection using a Small Cohort of MRI Images

arXiv cs.CV / 3/20/2026

📰 NewsIdeas & Deep AnalysisTools & Practical UsageModels & Research

共有:

Key Points

The paper addresses automatic prostate cancer detection on a small dataset of 162 T2-weighted MRI images using transfer learning and augmentation to mitigate data scarcity.
It benchmarks Vision Transformers (ViT, Swin), ResNet18, and classical methods (Logistic Regression, SVM, HOG+SVM); transfer-learned ResNet18 achieves the best metrics (90.9% accuracy, 95.2% sensitivity, AUC 0.905) with only 11M parameters.
Notably, handcrafted features (HOG+SVM) reach comparable accuracy (AUC 0.917), highlighting that simpler features can compete on small datasets.
Compared to state-of-the-art approaches requiring biparametric MRI and large cohorts, this method succeeds with only T2-weighted images, and a reader study shows AI outperforming radiologists (95.2% vs 67.5% sensitivity; Fleiss kappa 0.524); code and data are publicly available.

Abstract

Prostate cancer is a leading cause of mortality in men, yet interpretation of T2-weighted prostate MRI remains challenging due to subtle and heterogeneous lesions. We developed an interpretable framework for automatic cancer detection using a small dataset of 162 T2-weighted images (102 cancer, 60 normal), addressing data scarcity through transfer learning and augmentation. We performed a comprehensive comparison of Vision Transformers (ViT, Swin), CNNs (ResNet18), and classical methods (Logistic Regression, SVM, HOG+SVM). Transfer-learned ResNet18 achieved the best performance (90.9% accuracy, 95.2% sensitivity, AUC 0.905) with only 11M parameters, while Vision Transformers showed lower performance despite substantially higher complexity. Notably, HOG+SVM achieved comparable accuracy (AUC 0.917), highlighting the effectiveness of handcrafted features in small datasets. Unlike state-of-the-art approaches relying on biparametric MRI (T2+DWI) and large cohorts, our method achieves competitive performance using only T2-weighted images, reducing acquisition complexity and computational cost. In a reader study of 22 cases, five radiologists achieved a mean sensitivity of 67.5% (Fleiss Kappa = 0.524), compared to 95.2% for the AI model, suggesting potential for AI-assisted screening to reduce missed cancers and improve consistency. Code and data are publicly available.