Out-of-distribution multi-view auto-encoders for prostate cancer lesion detection

TL;DR

This paper introduces an out-of-distribution multi-view auto-encoder method for prostate cancer lesion detection in MRI, addressing data scarcity and improving detection accuracy by leveraging multiple imaging directions.

Contribution

It proposes a novel multi-stream auto-encoder approach for OOD detection in prostate MRI, enhancing lesion detection performance over single-view methods.

Findings

Achieved higher AUC (82.3) with multi-view approach compared to single view (73.1).

Demonstrated the effectiveness of OOD detection for prostate cancer MRI analysis.

Showed potential of OOD methods in medical imaging with limited annotated data.

Abstract

Traditional deep learning (DL) approaches based on supervised learning paradigms require large amounts of annotated data that are rarely available in the medical domain. Unsupervised Out-of-distribution (OOD) detection is an alternative that requires less annotated data. Further, OOD applications exploit the class skewness commonly present in medical data. Magnetic resonance imaging (MRI) has proven to be useful for prostate cancer (PCa) diagnosis and management, but current DL approaches rely on T2w axial MRI, which suffers from low out-of-plane resolution. We propose a multi-stream approach to accommodate different T2w directions to improve the performance of PCa lesion detection in an OOD approach. We evaluate our approach on a publicly available data-set, obtaining better detection results in terms of AUC when compared to a single direction approach (73.1 vs 82.3). Our results show the potential of OOD approaches for PCa lesion detection based on MRI.