Learning to Look Closer: A New Instance-Wise Loss for Small Cerebral Lesion Segmentation

2025/11/21 by Bouteille, Luc, Jaus, Alexander, Kleesiek, Jens +2
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #I.2.10 #I.4.6 #J.3

paper · doi:10.48550/arxiv.2511.17146

Abstract

Traditional loss functions in medical image segmentation, such as Dice, often under-segment small lesions because their small relative volume contributes negligibly to the overall loss. To address this, instance-wise loss functions and metrics have been proposed to evaluate segmentation quality on a per-lesion basis. We introduce CC-DiceCE, a loss function based on the CC-Metrics framework, and compare it with the existing blob loss. Both are benchmarked against a DiceCE baseline within the nnU-Net framework, which provides a robust and standardized setup. We find that CC-DiceCE loss increases detection (recall) with minimal to no degradation in segmentation performance, albeit at the cost of slightly more false positives. Furthermore, our multi-dataset study shows that CC-DiceCE generally outperforms blob loss.

Citations

• Every Component Counts: Rethinking the Measure of Success for Medical Semantic Segmentation in Multi-Instance Segmentation Tasks
• nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation
• The Brain Tumor Segmentation (BraTS-METS) Challenge 2023: Brain Metastasis Segmentation on Pre-treatment MRI
• Improving Segmentation of Objects with Varying Sizes in Biomedical Images using Instance-wise and Center-of-Instance Segmentation Loss Function
• Where is VALDO? VAscular Lesions Detection and segmentatiOn challenge at MICCAI 2021
• Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge
• The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

Related

• Improving Segmentation of Objects with Varying Sizes in Biomedical Images using Instance-wise and Center-of-Instance Segmentation Loss Function
• Universal Loss Reweighting to Balance Lesion Size Inequality in 3D\n Medical Image Segmentation
• Size and Smoothness Aware Adaptive Focal Loss for Small Tumor Segmentation
• Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation
• Tackling Bias in the Dice Similarity Coefficient: Introducing nDSC for White Matter Lesion Segmentation