FlowVN Trained on a Single Dataset Enables Rapid Reconstruction of Highly Accelerated 4D Flow MRI Across Multiple Sites
Magnetic Resonance in Medicine · 2026-07 · 96 (1) · 374-386
DOI: 10.1002/mrm.70317 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 26.02.2026. Verfasst von Sohaib Ayaz Qazi und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Imaging, Three-Dimensional, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The strategic research area in circulation and metabolism at Linköping University, ALF Grants, Region Östergötland, Vetenskapsrådet.
Abstract (Originalsprache)
Purpose
The aim of this study is to evaluate a deep variational network, FlowVN, for the reconstruction of heavily undersampled 4D Flow MRI across multiple sites.
Methods
FlowVN was trained on fully sampled 4D Flow MRI datasets of healthy volunteers from one site. The model was tested on retrospective undersampled data ( R = 6–22) of six normal volunteers from the same site and six from another site, prospectively undersampled data ( R = 12.4–13.8) of six healthy volunteers from the second site and six patients with aortic stenosis from first Site. Performance was evaluated using nRMSE, relative and angular error, average and maximum velocity, flow rate, volumetric flow, and turbulent kinetic energy (TKE), with a Wilcoxon signed‐rank test to assess the difference from ground truth.
Results
FlowVN showed minimal sensitivity to the number of training datasets and performed well even when trained on a single dataset. FlowVN also demonstrated good generalizability across sites. No significant difference in average and maximum velocity was observed up to R = 16. Total TKE was preserved up to R = 10 in the normal volunteers, but was well preserved even at higher acceleration factors in the aortic stenosis patients. Flow volumes through the ascending and descending aorta were well preserved for all acceleration factors, although ascending aorta flow volumes for data from one site were significantly different from ground truth for AF > 16.
Conclusion
FlowVN accurately reconstructs highly undersampled 4D Flow MRI from multiple sites using a model trained on a single dataset, maintaining excellent quantitative image quality even at very high acceleration factors.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden.
- Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden.
- Science for Life Laboratory, Linköping University, Linköping, Sweden.
ORCID-Profile: Sohaib Ayaz Qazi, Tino Ebbers, Petter Dyverfeldt
Förderung
- The strategic research area in circulation and metabolism at Linköping University
- ALF Grants, Region Östergötland
- Vetenskapsrådet
- Analytic Imaging Diagnostics Arena
Zitieren
APA
Qazi, S. A., Bianchessi, T., Viola, F., Trenti, C., Ylipää, E., Ebbers, T., & Dyverfeldt, P. (2026). FlowVN Trained on a Single Dataset Enables Rapid Reconstruction of Highly Accelerated 4D Flow MRI Across Multiple Sites. Magnetic Resonance in Medicine, 96(1), 374-386. https://doi.org/10.1002/mrm.70317
Vancouver
Qazi SA, Bianchessi T, Viola F, Trenti C, Ylipää E, Ebbers T, et al. FlowVN Trained on a Single Dataset Enables Rapid Reconstruction of Highly Accelerated 4D Flow MRI Across Multiple Sites. Magnetic Resonance in Medicine. 2026;96(1):374-386. doi:10.1002/mrm.70317
BibTeX
@article{Qazi2026,
author = {Qazi, Sohaib Ayaz and Bianchessi, Tamara and Viola, Federica and Trenti, Chiara and Ylipää, Erik and Ebbers, Tino and Dyverfeldt, Petter},
title = {FlowVN Trained on a Single Dataset Enables Rapid Reconstruction of Highly Accelerated 4D Flow MRI Across Multiple Sites},
journal = {Magnetic Resonance in Medicine},
year = {2026},
volume = {96},
number = {1},
pages = {374--386},
doi = {10.1002/mrm.70317}
}
EUVIMED