Topologically faithful multi-class segmentation in medical images
Authors:
Preprint
Abstract
Topological accuracy in medical image segmentation is a highly important property for downstream applications such as network analysis and flow modeling in vessels or cell counting. Recently, significant methodological advancements have brought well-founded concepts from algebraic topology to binary segmentation. However, these approaches have been underexplored in multi-class segmentation scenarios, where topological errors are common. We propose a general loss function for topologically faithful multi-class segmentation extending the recent Betti matching concept, which is based on induced matchings of persistence barcodes. We project the N-class segmentation problem to N single-class segmentation tasks, which allows us to use 1-parameter persistent homology making training of neural networks computationally feasible. We validate our method on a comprehensive set of four medical datasets with highly variant topological characteristics. Our loss formulation significantly enhances topological correctness in cardiac, cell, artery-vein, and Circle of Willis segmentation.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- OCTA-500: A retinal dataset for optical coherence tomography angiography studyMedical Image Analysis, 2024
- Structure-Preserving Instance Segmentation via Skeleton-Aware Distance Transform2023
- A Persistent Homology-Based Topological Loss for CNN-Based Multiclass Segmentation of CMRIEEE Transactions on Medical Imaging, 2022
- Loss odyssey in medical image segmentationMedical Image Analysis, 2021
- Dense cellular segmentation for EM using 2D–3D neural network ensemblesScientific Reports, 2021
- Cellular and Molecular Probing of Intact Human OrgansCell, 2020
- Beyond the Pixel-Wise Loss for Topology-Aware Delineation2018
- Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?IEEE Transactions on Medical Imaging, 2018
- Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations2017
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015