EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Semi-supervision for clinical contrast-weighted image synthesis from magnetic resonance fingerprinting

Mahmut Yurt, Cagan Alkan, Xiaozhi Cao, Congyu Liao, Zihan Zhou, Tolga Cukur, Ali Syed, John Pauly, Shreyas Vasanawala, Kawin Setsompop

Magnetic Resonance Materials in Physics, Biology and Medicine · 17.03.2026 · 39 (5) · 765-781

DOI: 10.1007/s10334-026-01338-7 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Magnetic Resonance Materials in Physics, Biology and Medicine, online erschienen am 17.03.2026. Verfasst von Mahmut Yurt und 9 weiteren Autor:innen. Beteiligte Einrichtung: Electrical Engineering, Stanford University, Stanford, CA, USA. myurt@stanford.edu. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Image Processing, Computer-Assisted. Die Arbeit nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NIMH NIH HHS, NIBIB NIH HHS, NICHD NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: To facilitate ease of data compilation across diverse populations for training models to synthesize clinical contrast-weighted images from magnetic resonance fingerprinting. METHODS: We leverage a semi-supervised training framework using highly accelerated acquisitions of the target contrasts used as ground truths. We utilize complementary randomized data sampling masks across training subjects and contrasts for homogeneous learning in k-space, together with multi-task learning. RESULTS: Our experiments indicate that the proposed method achieves high-quality synthesis with networks trained on retrospectively and prospectively undersampled data of the contrast-weighted images, enabling undersampling up to 12-16 × . CONCLUSIONS: The proposed method enables semi-supervised learning for synthesis from MRF with an end-to-end, ultra-fast training data acquisition protocol that is easier to obtain across a large population in clinical settings.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Magnetic Resonance ImagingImage Processing, Computer-AssistedHumansAlgorithmsContrast MediaSupervised Machine LearningImage Interpretation, Computer-AssistedBrainRetrospective Studies

Stichwörter der Autor:innen

Clinical contrastsMRIMagnetic resonance fingerprintingSemi-supervisionSynthesis

Substanzen

Contrast Media

Autor:innen und Institutionen

  • Electrical Engineering, Stanford University, Stanford, CA, USA. myurt@stanford.edu.
  • Electrical Engineering, Stanford University, Stanford, CA, USA.
  • Radiology, Stanford University, Stanford, CA, USA.
  • Radiology, University of California, San Francisco, San Francisco, CA, USA.
  • Electrical Engineering, Bilkent University, Ankara, Turkey.

Förderung

  • NIMH NIH HHS
  • NIBIB NIH HHS
  • NICHD NIH HHS
  • Eunice Kennedy Shriver National Institute of Child Health and Human Development
  • National Institute of Mental Health
  • National Institute of Biomedical Imaging and Bioengineering
  • GE Healthcare

Zitieren

APA

Yurt, M., Alkan, C., Cao, X., Liao, C., Zhou, Z., Cukur, T., Syed, A., Pauly, J., Vasanawala, S., & Setsompop, K. (2026). Semi-supervision for clinical contrast-weighted image synthesis from magnetic resonance fingerprinting. Magnetic Resonance Materials in Physics, Biology and Medicine, 39(5), 765-781. https://doi.org/10.1007/s10334-026-01338-7

Vancouver

Yurt M, Alkan C, Cao X, Liao C, Zhou Z, Cukur T, et al. Semi-supervision for clinical contrast-weighted image synthesis from magnetic resonance fingerprinting. Magnetic Resonance Materials in Physics, Biology and Medicine. 2026;39(5):765-781. doi:10.1007/s10334-026-01338-7

BibTeX
@article{Yurt2026,
  author = {Yurt, Mahmut and Alkan, Cagan and Cao, Xiaozhi and Liao, Congyu and Zhou, Zihan and Cukur, Tolga and Syed, Ali and Pauly, John and Vasanawala, Shreyas and Setsompop, Kawin},
  title = {Semi-supervision for clinical contrast-weighted image synthesis from magnetic resonance fingerprinting},
  journal = {Magnetic Resonance Materials in Physics, Biology and Medicine},
  year = {2026},
  volume = {39},
  number = {5},
  pages = {765--781},
  doi = {10.1007/s10334-026-01338-7}
}
Kostenlos starten →