Home > Publications Database > Preserving instance continuity and length in segmentation through connectivity-aware loss computation > print

001     281112
005     20250918102645.0
024 7 _ |a 10.48550/ARXIV.2509.03154
|2 doi
037 _ _ |a DZNE-2025-01073
100 1 _ |a Szustakowski, Karol
|0 P:(DE-HGF)0
|b 0
245 _ _ |a Preserving instance continuity and length in segmentation through connectivity-aware loss computation
260 _ _ |c 2025
|b arXiv
336 7 _ |a Preprint
|b preprint
|m preprint
|0 PUB:(DE-HGF)25
|s 1758110568_31854
|2 PUB:(DE-HGF)
336 7 _ |a WORKING_PAPER
|2 ORCID
336 7 _ |a Electronic Article
|0 28
|2 EndNote
336 7 _ |a preprint
|2 DRIVER
336 7 _ |a ARTICLE
|2 BibTeX
336 7 _ |a Output Types/Working Paper
|2 DataCite
520 _ _ |a In many biomedical segmentation tasks, the preservation of elongated structure continuity and length is more important than voxel-wise accuracy. We propose two novel loss functions, Negative Centerline Loss and Simplified Topology Loss, that, applied to Convolutional Neural Networks (CNNs), help preserve connectivity of output instances. Moreover, we discuss characteristics of experiment design, such as downscaling and spacing correction, that help obtain continuous segmentation masks. We evaluate our approach on a 3D light-sheet fluorescence microscopy dataset of axon initial segments (AIS), a task prone to discontinuity due to signal dropout. Compared to standard CNNs and existing topology-aware losses, our methods reduce the number of segmentation discontinuities per instance, particularly in regions with missing input signal, resulting in improved instance length calculation in downstream applications. Our findings demonstrate that structural priors embedded in the loss design can significantly enhance the reliability of segmentation for biological applications.
536 _ _ |a 351 - Brain Function (POF4-351)
|0 G:(DE-HGF)POF4-351
|c POF4-351
|f POF IV
|x 0
536 _ _ |a Helmholtz AI - Helmholtz Artificial Intelligence Coordination Unit – Local Unit FZJ (E.40401.62)
|0 G:(DE-Juel-1)E.40401.62
|c E.40401.62
|x 1
536 _ _ |a Helmholtz AI Consultant Team FB Information (E54.303.11)
|0 G:(DE-Juel-1)E54.303.11
|c E54.303.11
|x 2
588 _ _ |a Dataset connected to DataCite
650 _ 7 |a Computer Vision and Pattern Recognition (cs.CV)
|2 Other
650 _ 7 |a FOS: Computer and information sciences
|2 Other
650 _ 7 |a I.4.6; I.2.10
|2 Other
700 1 _ |a Frank, Luk
|0 P:(DE-2719)9003099
|b 1
700 1 _ |a Esser, Julia
|0 P:(DE-2719)2812406
|b 2
700 1 _ |a Gründemann, Jan
|0 P:(DE-2719)9001219
|b 3
700 1 _ |a Piraud, Marie
|0 P:(DE-HGF)0
|b 4
773 _ _ |a 10.48550/ARXIV.2509.03154
856 4 _ |u https://pub.dzne.de/record/281112/files/DZNE-2025-01073_Restricted.pdf
856 4 _ |u https://pub.dzne.de/record/281112/files/DZNE-2025-01073_Restricted.pdf?subformat=pdfa
|x pdfa
909 C O |o oai:pub.dzne.de:281112
|p VDB
910 1 _ |a Helmholtz Munich
|0 I:(DE-HGF)0
|b 0
|6 P:(DE-HGF)0
910 1 _ |a Helmholtz AI
|0 I:(DE-HGF)0
|b 0
|6 P:(DE-HGF)0
910 1 _ |a Deutsches Zentrum für Neurodegenerative Erkrankungen
|0 I:(DE-588)1065079516
|k DZNE
|b 1
|6 P:(DE-2719)9003099
910 1 _ |a Deutsches Zentrum für Neurodegenerative Erkrankungen
|0 I:(DE-588)1065079516
|k DZNE
|b 2
|6 P:(DE-2719)2812406
910 1 _ |a Deutsches Zentrum für Neurodegenerative Erkrankungen
|0 I:(DE-588)1065079516
|k DZNE
|b 3
|6 P:(DE-2719)9001219
910 1 _ |a Helmholtz Munich
|0 I:(DE-HGF)0
|b 4
|6 P:(DE-HGF)0
910 1 _ |a Helmholtz AI
|0 I:(DE-HGF)0
|b 4
|6 P:(DE-HGF)0
913 1 _ |a DE-HGF
|b Gesundheit
|l Neurodegenerative Diseases
|1 G:(DE-HGF)POF4-350
|0 G:(DE-HGF)POF4-351
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-300
|4 G:(DE-HGF)POF
|v Brain Function
|x 0
914 1 _ |y 2025
920 _ _ |l yes
920 1 _ |0 I:(DE-2719)5000069
|k AG Gründemann
|l Neural Circuit Computations
|x 0
980 _ _ |a preprint
980 _ _ |a VDB
980 _ _ |a I:(DE-2719)5000069
980 _ _ |a UNRESTRICTED

LibraryCollectionCLSMajorCLSMinorLanguageAuthor
Marc 21