Uncertainty maps highlight unreliable regions in segmentation predictions. However, most uncertainty evaluation metrics treat voxels independently, ignoring spatial context and anatomical structure. As a result, they may assign identical scores to qualitatively distinct patterns (e.g., scattered vs. boundary-aligned uncertainty). We propose three spatially aware metrics that incorporate structural and boundary information and conduct a thorough validation on medical imaging data from the prostate zonal segmentation challenge within the Medical Segmentation Decathlon. Our results demonstrate improved alignment with clinically important factors and better discrimination between meaningful and spurious uncertainty patterns.
@article{arxiv.2506.16589,
title = {Spatially-Aware Evaluation of Segmentation Uncertainty},
author = {Tal Zeevi and Eléonore V. Lieffrig and Lawrence H. Staib and John A. Onofrey},
journal= {arXiv preprint arXiv:2506.16589},
year = {2025}
}
Comments
Presented at the 4th Workshop on Uncertainty Quantification for Computer Vision (CVPR 2025), June 11, 2025. This version is not included in the official proceedings