EUVIMEDScopri EUVIMED
Open access · CC BY 4.0

Self‐calibrated B 1 + and B 0 Field Inhomogeneities Estimation ( SAFE ) in 3D Magnetic Resonance Fingerprinting

Mengze Gao iD, Xiaozhi Cao iD, Daniel Raz Abraham, Zihan Zhou, Nan Wang, Congyu Liao iD, Yang Yang, Hanlin Sun, Hongwei Li, Yan Bai, Yu Shen, Yaping Wu and 2 more

Magnetic Resonance in Medicine · 2026

DOI: 10.1002/mrm.70525 ↗

At a glance

Journal article in Magnetic Resonance in Medicine, 2026. Written by Mengze Gao and 13 co-authors. Institution: Department of Radiology, Stanford University, Stanford, California, USA. Major MEDLINE topics: Magnetic Resonance Imaging, Brain, Imaging, Three-Dimensional, Image Processing, Computer-Assisted. 26 references. The full text is freely available (open access, CC BY 4.0). Funded by: NIH HHS, National Institutes of Health.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Purpose

To develop SAFE, a self‐calibrated framework to estimate B 1 + and B 0 field inhomogeneities directly from conventional magnetic resonance fingerprinting (MRF) acquisitions and improve its quantification accuracy.

Methods

SAFE utilized a two‐step approach. First, two physics‐informed image markers are extracted from the MRF data to create a magnitude and a phase image that are highly correlated with B 1 + and B 0 inhomogeneities, respectively. Second, a deep learning (DL) network is applied to map these markers to quantitative field maps. SAFE was tested on 3D Spiral Projection Imaging MRF brain acquisition at 3T, where the network was trained and validated across a multi‐site, multi‐vendor dataset ( N = 358) from healthy volunteers and clinical population. The capability of SAFE to be applied to unseen MRF sequences with different signal preparations and flip‐angle trains through adaptively retraining without the need for additional training data was also tested.

Results

SAFE achieved normalized‐root‐mean‐square‐error within 3% against gold‐standard field‐calibration scans on both B 1 + and B 0 maps ( N = 32), with high performance remaining on datasets from scanners that the training data were acquired from. T1 and T2 biases were shown to be effectively corrected on healthy volunteers, a patient with brain tumor, and a pediatric subject. Tissue quantification accuracy after SAFE‐estimated field correction was validated on a large patient cohort ( N = 86). SAFE was also demonstrated to be adaptable to different MRF sequences without acquiring additional training data.

Conclusion

By combining physics‐informed image markers with DL, the proposed SAFE framework enables calibration‐free estimation of B 1 + and B 0 field inhomogeneities at 3T for whole brain MRF.

Subjects

MeSH terms (MEDLINE)

Magnetic Resonance ImagingBrainImaging, Three-DimensionalImage Processing, Computer-AssistedHumansDeep LearningMaleAdultAlgorithmsFemaleCalibrationHealthy VolunteersYoung AdultMiddle Aged

Author keywords

MR fingerprintingdeep learningfield inhomogeneityquantitative MRI

Authors and institutions

  • Department of Radiology, Stanford University, Stanford, California, USA.
  • Department of Electrical Engineering, Stanford University, Stanford, California, USA.
  • Department of Radiology, University of California San Francisco, San Francisco, California, USA.
  • Department of Radiology, The Third Hospital of Mianyang, Mianyang, Sichuan, China.
  • Sichuan Mental Health Center, Mianyang, Sichuan, China.
  • Department of Radiology, Henan Provincial People's Hospital, Zhengzhou, Henan, China.

ORCID profiles: Mengze Gao, Xiaozhi Cao, Congyu Liao

Funding

  • NIH HHS
  • National Institutes of Health

Cite

APA

Gao, M., Cao, X., Abraham, D. R., Zhou, Z., Wang, N., Liao, C., Yang, Y., Sun, H., Li, H., Bai, Y., Shen, Y., Wu, Y., Wang, M., & Setsompop, K. (2026). Self‐calibrated B 1 + and B 0 Field Inhomogeneities Estimation ( SAFE ) in 3D Magnetic Resonance Fingerprinting. Magnetic Resonance in Medicine. https://doi.org/10.1002/mrm.70525

Vancouver

Gao M, Cao X, Abraham DR, Zhou Z, Wang N, Liao C, et al. Self‐calibrated B 1 + and B 0 Field Inhomogeneities Estimation ( SAFE ) in 3D Magnetic Resonance Fingerprinting. Magnetic Resonance in Medicine. 2026. doi:10.1002/mrm.70525

BibTeX
@article{Gao2026,
  author = {Gao, Mengze and Cao, Xiaozhi and Abraham, Daniel Raz and Zhou, Zihan and Wang, Nan and Liao, Congyu and Yang, Yang and Sun, Hanlin and Li, Hongwei and Bai, Yan and Shen, Yu and Wu, Yaping and Wang, Meiyun and Setsompop, Kawin},
  title = {Self‐calibrated B 1 + and B 0 Field Inhomogeneities Estimation ( SAFE ) in 3D Magnetic Resonance Fingerprinting},
  journal = {Magnetic Resonance in Medicine},
  year = {2026},
  doi = {10.1002/mrm.70525}
}
Start for free →