Cross-Slice Attention and Evidential Critical Loss for Uncertainty-Aware Prostate Cancer Detection

TL;DR

This paper introduces a novel 2.5D cross-slice attention model with evidential critical loss for prostate cancer detection in MR images, improving accuracy and uncertainty estimation by leveraging volumetric information.

Contribution

The paper presents a new 2.5D attention model combined with evidential loss, enhancing prostate cancer detection and uncertainty quantification in MR imaging.

Findings

Achieved state-of-the-art detection performance on two datasets.

Improved epistemic uncertainty estimation.

Demonstrated effectiveness of the cross-slice attention mechanism.

Abstract

Current deep learning-based models typically analyze medical images in either 2D or 3D albeit disregarding volumetric information or suffering sub-optimal performance due to the anisotropic resolution of MR data. Furthermore, providing an accurate uncertainty estimation is beneficial to clinicians, as it indicates how confident a model is about its prediction. We propose a novel 2.5D cross-slice attention model that utilizes both global and local information, along with an evidential critical loss, to perform evidential deep learning for the detection in MR images of prostate cancer, one of the most common cancers and a leading cause of cancer-related death in men. We perform extensive experiments with our model on two different datasets and achieve state-of-the-art performance in prostate cancer detection along with improved epistemic uncertainty estimation. The implementation of the model is available at https://github.com/aL3x-O-o-Hung/GLCSA_ECLoss.