Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Boris Eberhardt1, Benedikt A Poser2, N Jon Shah3

  • 1Institute of Neuroscience and Medicine 4, Forschungszentrum Jülich, Jüich, Germany; RWTH Aachen University, Aachen, Germany.

Zeitschrift Fur Medizinische Physik
|February 11, 2022
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effects of riluzole on excitation-inhibition imbalance and functional connectivity in 22q11.2 deletion syndrome: a multimodal 7-Tesla MRI study.

Psychiatry research. Neuroimaging·2026
Same author

A novel distortion-matched anatomical imaging sequence for high-fidelity functional mapping in submillimeter-resolution fMRI.

Scientific reports·2026
Same author

Point/counterpoint: The use of perfusion-weighted MRI and amino acid PET for the identification of treatment-related changes.

Neuro-oncology·2026
Same author

Improved random coincidence estimation including triple coincidence detection in PET.

Frontiers in nuclear medicine·2026
Same author

Impact of thermal and physiological denoising on laminar functional connectivity.

Scientific reports·2026
Same author

Future prospects for image-derived input function in molecular imaging quantification with the 7 T MR-BrainPET insert.

Frontiers in neuroscience·2026

This study introduces a machine learning approach using generative adversarial networks (GANs) to reduce radiofrequency (RF) calibration time in ultra-high field MRI. This method achieves comparable excitation homogeneity to existing techniques while significantly cutting down scan duration.

Area of Science:

  • Magnetic Resonance Imaging (MRI)
  • Medical Physics
  • Artificial Intelligence in Medicine

Background:

  • Ultra-high field MRI (UHF-MRI) faces B1+ inhomogeneities due to short RF wavelengths in tissues.
  • Current methods for parallel transmit (pTX) excitation include full field map acquisition or robust pre-computed pulses, both with limitations.

Purpose of the Study:

  • To develop and evaluate a novel intermediate method for RF pulse design in UHF-MRI.
  • To reduce RF pulse calibration time while maintaining excitation homogeneity.
  • To leverage machine learning for synthesizing missing field map data.

Main Methods:

  • Utilized generative adversarial networks (GANs) for image-to-image translation to synthesize B1+ field maps.
  • Employed a subset of acquired field maps combined with GAN-synthesized maps for RF pulse design.
Keywords:
MRIMachine learningParallel transmissionRF pulse design

Related Experiment Videos

  • Investigated a predictive machine learning model for non-linear RF pulse design.

    • Main Results:

    • Achieved excitation homogeneity comparable to state-of-the-art methods using only half acquired and half synthesized B1+ maps.
    • Demonstrated a significant reduction in total calibration time.
    • Showcased the potential of GANs for deducing missing field map information.

    Conclusions:

    • The proposed machine learning approach offers a faster and efficient alternative for RF pulse calibration in UHF-MRI.
    • Combining acquired and synthesized field maps with ML models provides tailored excitation with reduced calibration overhead.
    • This method holds promise for improving the practicality and efficiency of UHF-MRI protocols.