DTU-Net: Learning Topological Similarity for Curvilinear Structure Segmentation
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Documents
- Pre-print
Submitted manuscript, 4.34 MB, PDF document
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.
Original language | English |
---|---|
Title of host publication | Information Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings |
Editors | Alejandro Frangi, Marleen de Bruijne, Demian Wassermann, Nassir Navab |
Number of pages | 13 |
Publisher | Springer |
Publication date | 2023 |
Pages | 654-666 |
ISBN (Print) | 978-3-031-34047-5 |
ISBN (Electronic) | 978-3-031-34048-2 |
DOIs | |
Publication status | Published - 2023 |
Event | 28th International Conference on Information Processing in Medical Imaging, IPMI 2023 - San Carlos de Bariloche, Argentina Duration: 18 Jun 2023 → 23 Jun 2023 |
Conference
Conference | 28th International Conference on Information Processing in Medical Imaging, IPMI 2023 |
---|---|
Land | Argentina |
By | San Carlos de Bariloche |
Periode | 18/06/2023 → 23/06/2023 |
Series | Lecture Notes in Computer Science |
---|---|
Volume | 13939 |
ISSN | 0302-9743 |
Bibliographical note
Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
- Curvilinear segmentation, topology preservation, triplet loss
Research areas
ID: 366267126