On Image Segmentation With Noisy Labels: Characterization and Volume Properties of the Optimal Solutions to Accuracy and Dice
This work addresses segmentation metric reliability in medical imaging, but it is incremental as it builds on existing metrics without introducing new methods.
The paper tackles the problem of medical image segmentation with noisy labels by analyzing the optimal solutions for Accuracy and Dice metrics, showing that their volumes can deviate significantly from the expected target volume and that Accuracy solutions have smaller or equal volumes compared to Dice solutions.
We study two of the most popular performance metrics in medical image segmentation, Accuracy and Dice, when the target labels are noisy. For both metrics, several statements related to characterization and volume properties of the set of optimal segmentations are proved, and associated experiments are provided. Our main insights are: (i) the volume of the solutions to both metrics may deviate significantly from the expected volume of the target, (ii) the volume of a solution to Accuracy is always less than or equal to the volume of a solution to Dice and (iii) the optimal solutions to both of these metrics coincide when the set of feasible segmentations is constrained to the set of segmentations with the volume equal to the expected volume of the target.