A Unified Loss for Handling Inter-Class and Intra-Class Imbalance in Medical Image Segmentation
Authors:
Open Access
- 11 April 2025
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (8) , 8842-8850
Abstract
In utilizing deep learning techniques for medical image segmentation, two types of imbalance issues are observed: inter-class imbalance between majority and minority classes and intra-class imbalance between easy and hard samples. However, existing loss functions typically confuse these issues, leading to enhancements that cater to only one aspect. Moreover, loss functions optimized for specific tasks often exhibit limited generalizability. To address these issues, we propose Inter-class and Intra-class Balance loss, as well as a unified loss termed Balance loss. The Inter-class Balance loss controls the extent of hard sample mining for majority class samples by considering the frequency of minority classes present in each input image. This approach requires no manual adjustment weights and adapts automatically to different datasets. The Intra-class Balance loss enhances the network's ability to learn from hard samples by performing mining on hard samples within each class. We evaluate our loss functions on five segmentation tasks with varying degrees of class imbalance. The experimental results show that our proposed Balance loss enhances segmentation performance compared with the current loss functions and exhibits superior robustness.Keywords
This publication has 11 references indexed in Scilit:
- Medical image synthesis via conditional GANs: Application to segmenting brain tumoursComputers in Biology and Medicine, 2024
- Do we really need dice? The hidden region-size biases of segmentation lossesMedical Image Analysis, 2023
- Data augmentation for medical imaging: A systematic literature reviewComputers in Biology and Medicine, 2022
- Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentationComputerized Medical Imaging and Graphics, 2022
- Image segmentation evaluation: a survey of methodsArtificial Intelligence Review, 2020
- Combo loss: Handling input and output imbalance in multi-organ segmentationComputerized Medical Imaging and Graphics, 2019
- Deep Q Learning Driven CT Pancreas Segmentation With Geometry-Aware U-NetIEEE Transactions on Medical Imaging, 2019
- A Three-Stage Deep Learning Model for Accurate Retinal Vessel SegmentationIEEE Journal of Biomedical and Health Informatics, 2018
- Learning from class-imbalanced data: Review of methods and applicationsExpert Systems with Applications, 2017
- WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physiciansComputerized Medical Imaging and Graphics, 2015