CRISP - Reliable Uncertainty Estimation for Medical Image Segmentation
This addresses the critical need for accurate uncertainty estimation in medical imaging, offering a domain-specific improvement over existing methods.
The paper tackles the problem of unreliable uncertainty estimation in medical image segmentation by proposing CRISP, a contrastive method that learns a joint latent space to encode anatomical priors, resulting in superior performance on four medical image databases compared to state-of-the-art approaches.
Accurate uncertainty estimation is a critical need for the medical imaging community. A variety of methods have been proposed, all direct extensions of classification uncertainty estimations techniques. The independent pixel-wise uncertainty estimates, often based on the probabilistic interpretation of neural networks, do not take into account anatomical prior knowledge and consequently provide sub-optimal results to many segmentation tasks. For this reason, we propose CRISP a ContRastive Image Segmentation for uncertainty Prediction method. At its core, CRISP implements a contrastive method to learn a joint latent space which encodes a distribution of valid segmentations and their corresponding images. We use this joint latent space to compare predictions to thousands of latent vectors and provide anatomically consistent uncertainty maps. Comprehensive studies performed on four medical image databases involving different modalities and organs underlines the superiority of our method compared to state-of-the-art approaches.