Encoding Clinical Priori in 3D Convolutional Neural Networks for Prostate Cancer Detection in bpMRI

TL;DR

This paper introduces a probabilistic clinical prior into 3D CNNs for prostate cancer detection in bpMRI, significantly improving diagnosis accuracy and lesion detection performance.

Contribution

It presents a novel method to encode clinical priors into CNN architectures, enhancing prostate cancer detection in medical imaging.

Findings

Up to 8.70% increase in AUROC for diagnosis

Average 1.08 pAUC improvement in lesion detection

Effective across multiple CNN architectures

Abstract

We hypothesize that anatomical priors can be viable mediums to infuse domain-specific clinical knowledge into state-of-the-art convolutional neural networks (CNN) based on the U-Net architecture. We introduce a probabilistic population prior which captures the spatial prevalence and zonal distinction of clinically significant prostate cancer (csPCa), in order to improve its computer-aided detection (CAD) in bi-parametric MR imaging (bpMRI). To evaluate performance, we train 3D adaptations of the U-Net, U-SEResNet, UNet++ and Attention U-Net using 800 institutional training-validation scans, paired with radiologically-estimated annotations and our computed prior. For 200 independent testing bpMRI scans with histologically-confirmed delineations of csPCa, our proposed method of encoding clinical priori demonstrates a strong ability to improve patient-based diagnosis (upto 8.70% increase in AUROC) and lesion-level detection (average increase of 1.08 pAUC between 0.1-10 false positives per patient) across all four architectures.