Enforcing Connectivity of 3D Linear Structures Using Their 2D Projections

Doruk Oner*, Hussein Osman, Mateusz Koziński, Pascal Fua

*Korrespondierende/r Autor/-in für diese Arbeit

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung

Abstract

Many biological and medical tasks require the delineation of 3D curvilinear structures such as blood vessels and neurites from image volumes. This is typically done using neural networks trained by minimizing voxel-wise loss functions that do not capture the topological properties of these structures. As a result, the connectivity of the recovered structures is often wrong, which lessens their usefulness. In this paper, we propose to improve the 3D connectivity of our results by minimizing a sum of topology-aware losses on their 2D projections. This suffices to increase the accuracy and to reduce the annotation effort required to provide the required annotated training data
Originalspracheenglisch
TitelMedical Image Computing and Computer Assisted Intervention – MICCAI 2022 - 25th International Conference, Proceedings
Redakteure/-innenLinwei Wang, Qi Dou, P. Thomas Fletcher, Stefanie Speidel, Shuo Li
ErscheinungsortCham
Herausgeber (Verlag)Springer
Seiten591–601
Seitenumfang11
ISBN (Print)9783031164422
DOIs
PublikationsstatusVeröffentlicht - 2022
Veranstaltung25th International Conference on Medical Image Computing and Computer Assisted Intervention: MICCAI 2022 - Singapur, Singapur
Dauer: 18 Sep. 202222 Sep. 2022

Publikationsreihe

NameLecture Notes in Computer Science
Band13435
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Konferenz

Konferenz25th International Conference on Medical Image Computing and Computer Assisted Intervention
KurztitelMICCAI 2022
Land/GebietSingapur
OrtSingapur
Zeitraum18/09/2222/09/22

ASJC Scopus subject areas

  • Theoretische Informatik
  • Informatik (insg.)

Fingerprint

Untersuchen Sie die Forschungsthemen von „Enforcing Connectivity of 3D Linear Structures Using Their 2D Projections“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren