Multi-Region Probabilistic Dice Similarity Coefficient using the Aitchison Distance and Bipartite Graph Matching
This work addresses the need for better validation metrics in medical imaging or computer vision, but it is incremental as it builds on existing DSC methods.
The authors tackled the problem of validating multi-region probabilistic image segmentations by extending the Dice similarity coefficient to handle unordered labels and uncertainty, using bipartite graph matching and two distance functions (absolute probability differences and Aitchison distance) to provide a robust accuracy measure.
Validation of image segmentation methods is of critical importance. Probabilistic image segmentation is increasingly popular as it captures uncertainty in the results. Image segmentation methods that support multi-region (as opposed to binary) delineation are more favourable as they capture interactions between the different objects in the image. The Dice similarity coefficient (DSC) has been a popular metric for evaluating the accuracy of automated or semi-automated segmentation methods by comparing their results to the ground truth. In this work, we develop an extension of the DSC to multi-region probabilistic segmentations (with unordered labels). We use bipartite graph matching to establish label correspondences and propose two functions that extend the DSC, one based on absolute probability differences and one based on the Aitchison distance. These provide a robust and accurate measure of multi-region probabilistic segmentation accuracy.