DTU-Net: Learning Topological Similarity for Curvilinear Structure Segmentation

Manxi Lin*, Kilian Zepf, Anders Nymark Christensen, Zahra Bashir, Morten Bo Søndergaard Svendsen, Martin Tolsgaard, Aasa Feragen*

*Corresponding author af dette arbejde
11 Citationer (Scopus)

Abstract

Curvilinear structure segmentation is important in medical imaging, quantifying structures such as vessels, airways, neurons, or organ boundaries in 2D slices. Segmentation via pixel-wise classification often fails to capture the small and low-contrast curvilinear structures. Prior topological information is typically used to address this problem, often at an expensive computational cost, and sometimes requiring prior knowledge of the expected topology. We present DTU-Net, a data-driven approach to topology-preserving curvilinear structure segmentation. DTU-Net consists of two sequential, lightweight U-Nets, dedicated to texture and topology, respectively. While the texture net makes a coarse prediction using image texture information, the topology net learns topological information from the coarse prediction by employing a triplet loss trained to recognize false and missed splits in the structure. We conduct experiments on a challenging multi-class ultrasound scan segmentation dataset as well as a well-known retinal imaging dataset. Results show that our model outperforms existing approaches in both pixel-wise segmentation accuracy and topological continuity, with no need for prior topological knowledge.

OriginalsprogEngelsk
TitelInformation Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings
RedaktørerAlejandro Frangi, Marleen de Bruijne, Demian Wassermann, Nassir Navab
Antal sider13
ForlagSpringer Science and Business Media Deutschland GmbH
Publikationsdato2023
Sider654-666
ISBN (Trykt)9783031340475
DOI
StatusUdgivet - 2023
Begivenhed28th International Conference on Information Processing in Medical Imaging, IPMI 2023 - San Carlos de Bariloche, Argentina
Varighed: 18 jun. 202323 jun. 2023

Konference

Konference28th International Conference on Information Processing in Medical Imaging, IPMI 2023
Land/OmrådeArgentina
BySan Carlos de Bariloche
Periode18/06/202323/06/2023
NavnLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vol/bind13939 LNCS
ISSN0302-9743

Fingeraftryk

Dyk ned i forskningsemnerne om 'DTU-Net: Learning Topological Similarity for Curvilinear Structure Segmentation'. Sammen danner de et unikt fingeraftryk.

Citationsformater