EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Dynamic Mode Decomposition ( DMD ) for Low‐Latency Real‐Time Cardiac MRI

Ecrin Yagiz iD, Bilal Tasdelen iD, Ibrahim K Ozaslan, Mihailo R Jovanovic, Ye Tian iD, Krishna S Nayak iD

Magnetic Resonance in Medicine · 2026-08 · 96 (2) · 623-634

DOI: 10.1002/mrm.70360 ↗

At a glance

Journal article in Magnetic Resonance in Medicine, published online on 2026-03-30. Written by Ecrin Yagiz and 5 co-authors. Institution: Ming Hsieh Department of Electrical and Computer Engineering, Viterbi School of Engineering, University of Southern California, Los Angeles, Californ…. Major MEDLINE topics: Heart, Magnetic Resonance Imaging, Image Processing, Computer-Assisted, Image Interpretation, Computer-Assisted. 69 references. The full text is freely available (open access, CC BY 4.0). Funded by: NIH HHS, NICHD NIH HHS, NHLBI NIH HHS.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Purpose

To demonstrate dynamic mode decomposition ( DMD ) for high spatiotemporal low‐latency online reconstruction in 2D real‐time cardiac MRI.

Methods

DMD was applied to 2D spiral balanced steady state free precession ( bSSFP ) real‐time adult and fetal cardiac MRI at 0.55 T, with data from 10 healthy adult volunteers (3F/7M; age: 21–49; BMI: 20–34) and 6 pregnant females (maternal age: 30–41; maternal BMI: 22–47; gestational age: 23 weeks 6 days–37 weeks 5 days). DMD model appropriateness was assessed against off‐line spatiotemporally constrained reconstruction ( STCR ) as the reference. We retrospectively evaluated DMD‐based low‐latency online reconstruction at two temporal resolutions (21 and 42 ms/frame). DMD modes were estimated from the most recently acquired frames and used to remove aliasing while preserving underlying physiological motion.

Results

DMD represented cardiac dynamics with normalized root‐mean‐square error ( NRMSE ) less than 7% when all modes retained. Low‐latency DMD‐based online reconstruction performed de‐aliasing while preserving the physiological motion, supporting framerates (21 and 42 ms/frame).

Conclusion

We have demonstrated that the DMD framework is applicable to 2D real‐time cardiac MRI and for low‐latency de‐aliasing for better online reconstruction.

Subjects

MeSH terms (MEDLINE)

HeartMagnetic Resonance ImagingImage Processing, Computer-AssistedImage Interpretation, Computer-AssistedMagnetic Resonance Imaging, CineHumansFemaleAdultAlgorithmsPregnancyYoung AdultReproducibility of ResultsImage EnhancementRetrospective Studies

Author keywords

cardiac MRIdynamic mode decompositionimage reconstructioninteractive imagingonline reconstruction

Authors and institutions

  • Ming Hsieh Department of Electrical and Computer Engineering, Viterbi School of Engineering, University of Southern California, Los Angeles, California, USA.

ORCID profiles: Ecrin Yagiz, Bilal Tasdelen, Ye Tian, Krishna S Nayak

Funding

  • NIH HHS
  • NICHD NIH HHS
  • NHLBI NIH HHS
  • National Institutes of Health

Cite

APA

Yagiz, E., Tasdelen, B., Ozaslan, I. K., Jovanovic, M. R., Tian, Y., & Nayak, K. S. (2026). Dynamic Mode Decomposition ( DMD ) for Low‐Latency Real‐Time Cardiac MRI. Magnetic Resonance in Medicine, 96(2), 623-634. https://doi.org/10.1002/mrm.70360

Vancouver

Yagiz E, Tasdelen B, Ozaslan IK, Jovanovic MR, Tian Y, Nayak KS. Dynamic Mode Decomposition ( DMD ) for Low‐Latency Real‐Time Cardiac MRI. Magnetic Resonance in Medicine. 2026;96(2):623-634. doi:10.1002/mrm.70360

BibTeX
@article{Yagiz2026,
  author = {Yagiz, Ecrin and Tasdelen, Bilal and Ozaslan, Ibrahim K and Jovanovic, Mihailo R and Tian, Ye and Nayak, Krishna S},
  title = {Dynamic Mode Decomposition ( DMD ) for Low‐Latency Real‐Time Cardiac MRI},
  journal = {Magnetic Resonance in Medicine},
  year = {2026},
  volume = {96},
  number = {2},
  pages = {623--634},
  doi = {10.1002/mrm.70360}
}
Start for free →