Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy

D 2025

Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy

MELNIKOVA, Aleksandra a Petr MATULA

Základní údaje

Originální název

Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy

Autoři

MELNIKOVA, Aleksandra a Petr MATULA

Vydání

ISBI 2025, Proceedings - International Symposium on Biomedical Imaging, od s. 1-5, 5 s. 2025

Nakladatel

IEEE Computer Society

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10200 1.2 Computer and information sciences

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

URL

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14330/25:00144299

Organizační jednotka

Fakulta informatiky

ISBN

979-8-3315-2052-6

ISSN

Klíčová slova anglicky

Annotation fusion; benchmarking; cell segmentation; reference data; shape analysis

Štítky

cbia-web, firank_B

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 2. 4. 2026 14:31, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

In biomedical image processing, reference data used to benchmark segmentation algorithms are typically obtained by fusing expert annotations. We compare popular fusion algorithms (majority voting, STAPLE, and Robust Bayesian Fusion) and study (1) how their results depend on the variability of input annotations and (2) how well they preserve shapes w.r.t, biologically relevant measures (BRM) such as volume and average roundness. We also study (3) the correlation between BRM and measures common in image processing (namely, Jaccard index and Hausdorff distance). For our experiments, we created 3D time-lapse segmentation annotations of migrating cells with complex dynamic shapes (Fluo-C3DL-MDA231 dataset from the Cell Tracking Challenge). The full annotations are publicly available. The study found that none of the selected methods produced satisfactory results w.r.t. BRM, highlighting the need for new methods.

Návaznosti

GA21-20374S, projekt VaV

Název: Segmentace a sledování buněk se složitým tvarem

Investor: Grantová agentura ČR, Segmentace a sledování buněk se složitým tvarem

MUNI/A/1600/2024, interní kód MU

Název: Modelování, analýza a verifikace (2025)

Investor: Masarykova univerzita, Modelování, analýza a verifikace (2025)

Citovat

MELNIKOVA, Aleksandra a Petr MATULA. Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy. Online. In Proceedings - International Symposium on Biomedical Imaging. ISBI 2025: IEEE Computer Society, 2025, s. 1-5. ISBN 979-8-3315-2052-6. Dostupné z: https://doi.org/10.1109/ISBI60581.2025.10980908.

@inproceedings{2488125,
   author = {Melnikova, Aleksandra and Matula, Petr},
   address = {ISBI 2025},
   booktitle = {Proceedings - International Symposium on Biomedical Imaging},
   doi = {https://doi.org/10.1109/ISBI60581.2025.10980908},
   keywords = {Annotation fusion; benchmarking; cell segmentation; reference data; shape analysis},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {ISBI 2025},
   isbn = {979-8-3315-2052-6},
   pages = {1-5},
   publisher = {IEEE Computer Society},
   title = {Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy},
   url = {https://ieeexplore.ieee.org/document/10980908},
   year = {2025}
}

TY  - CONF
ID  - 2488125
AU  - Melnikova, Aleksandra - Matula, Petr
PY  - 2025
TI  - Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy
PB  - IEEE Computer Society
CY  - ISBI 2025
SN  - 9798331520526
KW  - Annotation fusion
KW  - benchmarking
KW  - cell segmentation
KW  - reference data
KW  - shape analysis
UR  - https://ieeexplore.ieee.org/document/10980908
N2  - In biomedical image processing, reference data used to benchmark segmentation algorithms are typically obtained by fusing expert annotations. We compare popular fusion algorithms (majority voting, STAPLE, and Robust Bayesian Fusion) and study (1) how their results depend on the variability of input annotations and (2) how well they preserve shapes w.r.t, biologically relevant measures (BRM) such as volume and average roundness. We also study (3) the correlation between BRM and measures common in image processing (namely, Jaccard index and Hausdorff distance). For our experiments, we created 3D time-lapse segmentation annotations of migrating cells with complex dynamic shapes (Fluo-C3DL-MDA231 dataset from the Cell Tracking Challenge). The full annotations are publicly available. The study found that none of the selected methods produced satisfactory results w.r.t. BRM, highlighting the need for new methods.
ER  -

MELNIKOVA, Aleksandra a Petr MATULA. Study of Shape Fusion Algorithms for 3D Time-Lapse Microscopy. Online. In \textit{Proceedings - International Symposium on Biomedical Imaging}. ISBI 2025: IEEE Computer Society, 2025, s.~1-5. ISBN~979-8-3315-2052-6. Dostupné z: https://doi.org/10.1109/ISBI60581.2025.10980908.

Přehled o publikaci