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 Video

Updated: Sep 8, 2025

Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
08:51

Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla

Published on: February 19, 2021

9.2K

EPI phase error correction with deep learning (PEC-DL) at 7 T.

Lili Wang1, Chengyan Wang2, Fanwen Wang1

  • 1Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, People's Republic of China.

Magnetic Resonance in Medicine
|June 13, 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

Multimodal Bidirectional Direct Preference Optimization and Instruction Fine-Tuning for Medical Image Understanding and Generation.

IEEE journal of biomedical and health informatics·2026
Same author

Stellate ganglion block attenuates gut barrier injury in sleep-deprived rats in a gut microbiota-dependent manner.

Scientific reports·2026
Same author

Peripheral blood IFN-γ-producing T-cell subsets and soluble IL-2 receptor as independent prognostic biomarkers in NSCLC treated with immune checkpoint inhibitor-based therapy.

Discover oncology·2026
Same author

Novel Metabolomic and Adiposity Markers Improve Prediction of Obstructive Sleep Apnea: A Prospective Analysis in the China Human Phenobank.

Phenomics (Cham, Switzerland)·2026
Same author

Correlation between functional impairment and quantitative MRI assessment of ankle cartilage in chronic ankle instability.

Asia-Pacific journal of sports medicine, arthroscopy, rehabilitation and technology·2026
Same author

RFD: A Reducing Feature Discrepancy method for unsupervised cross-modality SAM adaptation.

Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society·2026
Same journal

Multi-Contrast Human Brain CEST MRI at 11.7 T: First In Vivo Demonstration.

Magnetic resonance in medicine·2026
Same journal

Suppression of Oscillation and Ghosting in RF-Spoiled Gradient-Echo-Based Dynamic Imaging.

Magnetic resonance in medicine·2026
Same journal

A Simple, Dynamic Geometric Phantom for MRI and CT Reconstruction Pipelines: Beyond Shepp-Logan.

Magnetic resonance in medicine·2026
Same journal

7T 3D-EPI PCASL With High SNR Efficiency and Robustness to Through-Plane B<sub>0</sub> Field Gradients.

Magnetic resonance in medicine·2026
Same journal

A Comparison of Tissue Property Values Estimated Using Conventional Cardiac MRF and MT-Cardiac MRF.

Magnetic resonance in medicine·2026
Same journal

Dependence of the Extra-Cellular Diffusion Coefficient on the Fractions of Neurites and Cell Bodies in Gray Matter.

Magnetic resonance in medicine·2026
See all related articles

A novel deep learning method, PEC-DL, effectively corrects Nyquist ghost artifacts in echo-planar imaging (EPI) diffusion-weighted imaging (DWI) at 7 Tesla. This technique works even with accelerated data, significantly reducing artifacts without calibration.

Area of Science:

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

Background:

  • Nyquist ghost artifacts in EPI are caused by phase mismatches, degrading image quality.
  • Current ghost correction methods struggle with k-space undersampling and leave residual artifacts.
  • Diffusion-weighted imaging (DWI) at 7 Tesla offers high resolution but is susceptible to EPI artifacts.

Purpose of the Study:

  • To develop and validate a deep learning-based method (PEC-DL) for correcting phase errors and Nyquist ghost artifacts in 7 Tesla DWI.
  • To assess the efficacy of PEC-DL on both fully sampled and undersampled EPI data.
  • To evaluate PEC-DL's performance in healthy volunteers and patients with Moyamoya disease.

Main Methods:

  • A deep learning network (PEC-DL) was trained on fully sampled EPI data.
Keywords:
7 TeslaEPIdeep learningghostphase correction

More Related Videos

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.3K
Cardiac Magnetic Resonance Imaging at 7 Tesla
09:14

Cardiac Magnetic Resonance Imaging at 7 Tesla

Published on: January 6, 2019

11.6K

Related Experiment Videos

Last Updated: Sep 8, 2025

Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
08:51

Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla

Published on: February 19, 2021

9.2K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.3K
Cardiac Magnetic Resonance Imaging at 7 Tesla
09:14

Cardiac Magnetic Resonance Imaging at 7 Tesla

Published on: January 6, 2019

11.6K
  • Undersampled k-space data were divided into two sets based on readout polarity.
  • The PEC-DL network reconstructed ghost-free images from undersampled data without requiring calibration or navigator data.

    • Main Results:

    • PEC-DL successfully mitigated ghost artifacts in DWI for both healthy volunteers and Moyamoya disease patients.
    • Fourfold accelerated data showed reduced distortion in Moyamoya lesions on high b-value DWI and ADC maps.
    • PEC-DL achieved significantly lower ghost-to-signal ratios compared to conventional methods like linear phase correction, mini-entropy, and PEC-GRAPPA.

    Conclusions:

    • The proposed PEC-DL method effectively eliminates ghost artifacts in EPI data.
    • PEC-DL is applicable to both fully sampled and accelerated (up to fourfold) EPI data.
    • The method operates without the need for calibration or navigator data, simplifying its application.