Towards Reliable WMH Segmentation under Domain Shift: An Application Study using Maximum Entropy Regularization to Improve Uncertainty Estimation
This work addresses domain shift challenges in medical imaging for clinical applications like multiple sclerosis, but it is incremental as it builds on existing methods with specific improvements.
The study tackled the problem of domain shift in white matter hyperintensity segmentation by applying maximum-entropy regularization to improve model calibration and uncertainty estimation, showing that entropy-based uncertainty can predict segmentation errors and enhance correlation with performance under domain shift.
Accurate segmentation of white matter hyperintensities (WMH) is crucial for clinical decision-making, particularly in the context of multiple sclerosis. However, domain shifts, such as variations in MRI machine types or acquisition parameters, pose significant challenges to model calibration and uncertainty estimation. This study investigates the impact of domain shift on WMH segmentation by proposing maximum-entropy regularization techniques to enhance model calibration and uncertainty estimation, with the purpose of identifying errors post-deployment using predictive uncertainty as a proxy measure that does not require ground-truth labels. To do this, we conducted experiments using a U-Net architecture to evaluate these regularization schemes on two publicly available datasets, assessing performance with the Dice coefficient, expected calibration error, and entropy-based uncertainty estimates. Our results show that entropy-based uncertainty estimates can anticipate segmentation errors, and that maximum-entropy regularization further strengthens the correlation between uncertainty and segmentation performance while also improving model calibration under domain shift.