Uncertainty-Error correlations in Evidential Deep Learning models for biomedical segmentation

TL;DR

This paper evaluates Evidential Deep Learning models for biomedical image segmentation, demonstrating their superior uncertainty-error correlation and potential for improved active learning compared to traditional methods.

Contribution

It introduces the application of Evidential Deep Learning with Dirichlet priors to biomedical segmentation and compares its effectiveness against standard uncertainty measures.

Findings

EDL models outperform traditional methods in uncertainty-error correlation.

EDL models achieve similar segmentation accuracy with better uncertainty estimates.

Enhanced active learning performance with EDL models in biomedical segmentation.

Abstract

In this work, we examine the effectiveness of an uncertainty quantification framework known as Evidential Deep Learning applied in the context of biomedical image segmentation. This class of models involves assigning Dirichlet distributions as priors for segmentation labels, and enables a few distinct definitions of model uncertainties. Using the cardiac and prostate MRI images available in the Medical Segmentation Decathlon for validation, we found that Evidential Deep Learning models with U-Net backbones generally yielded superior correlations between prediction errors and uncertainties relative to the conventional baseline equipped with Shannon entropy measure, Monte-Carlo Dropout and Deep Ensemble methods. We also examined these models' effectiveness in active learning, finding that relative to the standard Shannon entropy-based sampling, they yielded higher point-biserial uncertainty-error correlations while attaining similar performances in Dice-Sorensen coefficients. These superior features of EDL models render them well-suited for segmentation tasks that warrant a critical sensitivity in detecting large model errors.