Topology Preserving

Topology-preserving segmentation aims to accurately delineate objects in images while maintaining their inherent connectivity and structure, a crucial aspect often overlooked by standard segmentation methods. Current research focuses on developing novel loss functions and network architectures, such as deformation-based models incorporating quasiconformal mappings or distance transforms, to enforce topological constraints during the segmentation process. These advancements improve the accuracy and reliability of segmentation in various applications, particularly in medical imaging (e.g., airway or vessel segmentation) and other fields requiring precise structural representation, leading to better downstream analysis and clinical decision-making.

Papers

November 5, 2024

Topograph: An efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation
Laurin Lux, Alexander H. Berger, Alexander Weers, Nico Stucki, Daniel Rueckert, Ulrich Bauer, Johannes C. Paetzold
Topology Problem Topological Information Topological Accuracy Topology Preserving Graph Based Framework

April 3, 2024

Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures
Yannick Kirchhoff, Maximilian R. Rokuss, Saikat Roy, Balint Kovacs, Constantin Ulrich, Tassilo Wald, Maximilian Zenk, Philipp Vollmuth, Jens Kleesiek, Fabian Isensee, Klaus Maier-Hein
Efficient Segmentation Segmentation Loss Tubular Structure Topology Aware Topology Preserving

October 7, 2022

A Learning-based Framework for Topology-Preserving Segmentation using Quasiconformal Mappings
Han Zhang, Lok Ming Lui
Segmentation Based Approach Segmentation Mask Learning Framework Better Segmentation Conformal Mapping Multi Object Segmentation Semantic Aware Deformation Topology Preserving

September 17, 2022

Differentiable Topology-Preserved Distance Transform for Pulmonary Airway Segmentation
Minghui Zhang, Guang-Zhong Yang, Yun Gu
Closed Form Differentiable Expression Airway Segmentation Preserving Transformation Efficient Surrogate Model Pulmonary Airway Segmentation Topology Preserving

July 23, 2022

BuyTheDips: PathLoss for improved topology-preserving deep learning-based image segmentation
Minh On Vu Ngoc, Yizi Chen, Nicolas Boutry, Jonathan Fabrizio, Clement Mallet
Accurate Segmentation Path Loss Topological Loss Topology Preserving

February 27, 2022

Topology-Preserving Segmentation Network: A Deep Learning Segmentation Framework for Connected Component
Han Zhang, Lok Ming Lui
Learning Based Segmentation Network Segmentation Accuracy Connected Component Topology Preserving

Topology Preserving

Papers

Topograph: An efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation

Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures

A Learning-based Framework for Topology-Preserving Segmentation using Quasiconformal Mappings

Differentiable Topology-Preserved Distance Transform for Pulmonary Airway Segmentation

BuyTheDips: PathLoss for improved topology-preserving deep learning-based image segmentation

Topology-Preserving Segmentation Network: A Deep Learning Segmentation Framework for Connected Component