EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access

Fine‐Tuning Deep Learning Model for Quantitative Knee Joint Mapping With MR Fingerprinting and Its Comparison to Dictionary Matching Method

Xiaoxia Zhang iD, Hector L de Moura iD, Anmol Monga iD, Marcelo V W Zibetti iD, Ravinder R Regatte iD

NMR in Biomedicine · 2025

DOI: 10.1002/nbm.70045 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift NMR in Biomedicine, 2025. Verfasst von Xiaoxia Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA. Hauptthemen laut MEDLINE: Deep Learning, Magnetic Resonance Imaging, Knee Joint. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIH HHS, NIAMS NIH HHS, NIBIB NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Magnetic resonance fingerprinting (MRF), as an emerging versatile and noninvasive imaging technique, provides simultaneous quantification of multiple quantitative MRI parameters, which have been used to detect changes in cartilage composition and structure in osteoarthritis. Deep learning (DL)–based methods for quantification mapping in MRF overcome the memory constraints and offer faster processing compared to the conventional dictionary matching (DM) method. However, limited attention has been given to the fine‐tuning of neural networks (NNs) in DL and fair comparison with DM. In this study, we investigate the impact of training parameter choices on NN performance and compare the fine‐tuned NN with DM for multiparametric mapping in MRF. Our approach includes optimizing NN hyperparameters, analyzing the singular value decomposition (SVD) components of MRF data, and optimization of the DM method. We conducted experiments on synthetic data, the NIST/ISMRM MRI system phantom with ground truth, and in vivo knee data from 14 healthy volunteers. The results demonstrate the critical importance of selecting appropriate training parameters, as these significantly affect NN performance. The findings also show that NNs improve the accuracy and robustness of T 1 , T 2 , and T 1ρ mappings compared to DM in synthetic datasets. For in vivo knee data, the NN achieved comparable results for T 1 , with slightly lower T 2 and slightly higher T 1ρ measurements compared to DM. In conclusion, the fine‐tuned NN can be used to increase accuracy and robustness for multiparametric quantitative mapping from MRF of the knee joint.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningMagnetic Resonance ImagingKnee JointHumansAdultFemaleMaleNeural Networks, ComputerPhantoms, Imaging

Stichwörter der Autor:innen

MR fingerprintingdeep learningmusculoskeletal imagingquantitative MRI

Autor:innen und Institutionen

  • Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA.

ORCID-Profile: Xiaoxia Zhang, Hector L de Moura, Anmol Monga, Marcelo V W Zibetti, Ravinder R Regatte

Förderung

  • NIH HHS
  • NIAMS NIH HHS
  • NIBIB NIH HHS
  • National Institutes of Health

Zitieren

APA

Zhang, X., de Moura, H. L., Monga, A., Zibetti, M. V. W., & Regatte, R. R. (2025). Fine‐Tuning Deep Learning Model for Quantitative Knee Joint Mapping With MR Fingerprinting and Its Comparison to Dictionary Matching Method. NMR in Biomedicine. https://doi.org/10.1002/nbm.70045

Vancouver

Zhang X, de Moura HL, Monga A, Zibetti MVW, Regatte RR. Fine‐Tuning Deep Learning Model for Quantitative Knee Joint Mapping With MR Fingerprinting and Its Comparison to Dictionary Matching Method. NMR in Biomedicine. 2025. doi:10.1002/nbm.70045

BibTeX
@article{Zhang2025,
  author = {Zhang, Xiaoxia and de Moura, Hector L and Monga, Anmol and Zibetti, Marcelo V W and Regatte, Ravinder R},
  title = {Fine‐Tuning Deep Learning Model for Quantitative Knee Joint Mapping With MR Fingerprinting and Its Comparison to Dictionary Matching Method},
  journal = {NMR in Biomedicine},
  year = {2025},
  doi = {10.1002/nbm.70045}
}
Kostenlos starten →