3D Segmentation of Multi-contrast Cardiac Magnetic Resonances With Topological Correction and Synthetic Data Augmentation
Annals of Biomedical Engineering · 2026
DOI: 10.1007/s10439-026-04115-8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Annals of Biomedical Engineering, 2026. Verfasst von Ricardo M Rosales und 2 weiteren Autor:innen. Beteiligte Einrichtung: Aragón Institute of Engineering Research (I3A), Zaragoza, Aragón, Spain. rrosales@unizar.es. Hauptthemen laut MEDLINE: Imaging, Three-Dimensional, Heart, Magnetic Resonance Imaging. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Horizon 2020 Framework Programme, Agencia Estatal de Investigación, Gobierno de Aragón.
Abstract (Originalsprache)
Purpose
Automatic segmentation of cardiac magnetic resonance (CMR) images improves the evaluation of heart structure and function, helping clinical diagnosis and the generation of in silico models. Recent advances have introduced synthetic augmentation (SA) using generative adversarial networks (GANs) and topological correction (TC) via persistent homology to enhance segmentation with convolutional neural networks (CNNs). However, their combined effectiveness remains unexplored. Here, we extend and systematically evaluate these techniques, individually and in combination, for the first time in the context of three-dimensional (3D) CMR segmentation across challenging multi-vendor, multi-center, multi-class, and multi-contrast data sets.
Methods
Data involved anisotropic, topologically inconsistent cine and late gadolinium-enhanced (LGE) CMRs, and isotropic, topologically consistent ex vivo CMRs. Topological priors were defined in each data set from ground truth label (GTL) assessments, and TC was applied by retraining the baseline 3D CNN with a loss function accounting for topological discrepancies. For SA, deformed GTLs were used to generate synthetic images using trained 3D GANs.
Results
Consistent segmentation improvements were observed for the ex vivo data in both overlap with GTLs and topological precision when applying TC and SA individually and in combination. Notably, an enhanced identification of the infarction was obtained when SA and TC were used in the LGE data. Overall, SA increased the predictions overlap with GTLs, while TC reduced the topological discrepancies across all data sets.
Conclusion
TC and SA demonstrate strong potential for improving 3D CMR segmentation on complex, real-world data sets, especially when topologically consistent data are available for training.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Aragón Institute of Engineering Research (I3A), Zaragoza, Aragón, Spain. rrosales@unizar.es.
- Aragón Institute for Health Research (IISA), Zaragoza, Aragón, Spain. rrosales@unizar.es.
- University of Zaragoza, Zaragoza, Aragón, Spain. rrosales@unizar.es.
- Aragón Institute of Engineering Research (I3A), Zaragoza, Aragón, Spain.
- Aragón Institute for Health Research (IISA), Zaragoza, Aragón, Spain.
- University of Zaragoza, Zaragoza, Aragón, Spain.
Förderung
- Horizon 2020 Framework Programme
- Agencia Estatal de Investigación
- Gobierno de Aragón
- H2020 European Research Council
- Universidad de Zaragoza
Zitieren
APA
Rosales, R. M., Doblaré, M., & Pueyo, E. (2026). 3D Segmentation of Multi-contrast Cardiac Magnetic Resonances With Topological Correction and Synthetic Data Augmentation. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04115-8
Vancouver
Rosales RM, Doblaré M, Pueyo E. 3D Segmentation of Multi-contrast Cardiac Magnetic Resonances With Topological Correction and Synthetic Data Augmentation. Annals of Biomedical Engineering. 2026. doi:10.1007/s10439-026-04115-8
BibTeX
@article{Rosales2026,
author = {Rosales, Ricardo M and Doblaré, Manuel and Pueyo, Esther},
title = {3D Segmentation of Multi-contrast Cardiac Magnetic Resonances With Topological Correction and Synthetic Data Augmentation},
journal = {Annals of Biomedical Engineering},
year = {2026},
doi = {10.1007/s10439-026-04115-8}
}
EUVIMED