TY - GEN

T1 - DTU-Net

T2 - 28th International Conference on Information Processing in Medical Imaging, IPMI 2023

AU - Lin, Manxi

AU - Zepf, Kilian

AU - Christensen, Anders Nymark

AU - Bashir, Zahra

AU - Svendsen, Morten Bo Søndergaard

AU - Tolsgaard, Martin

AU - Feragen, Aasa

N1 - Publisher Copyright: © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

PY - 2023

Y1 - 2023

N2 - 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.

AB - 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.

KW - Curvilinear segmentation

KW - topology preservation

KW - triplet loss

U2 - 10.1007/978-3-031-34048-2_50

DO - 10.1007/978-3-031-34048-2_50

M3 - Article in proceedings

AN - SCOPUS:85164012711

SN - 978-3-031-34047-5

T3 - Lecture Notes in Computer Science

SP - 654

EP - 666

BT - Information Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings

A2 - Frangi, Alejandro

A2 - de Bruijne, Marleen

A2 - Wassermann, Demian

A2 - Navab, Nassir

PB - Springer

Y2 - 18 June 2023 through 23 June 2023

ER -