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 Concept Videos

Deconvolution01:20

Deconvolution

460
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
460

You might also read

Related Articles

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

Sort by
Same author

Imaging Near Spinal Fixation Hardware at 0.55 T Compared With 3 T.

Journal of magnetic resonance imaging : JMRI·2026
Same author

Ghost poisoning: Making users invisible to speaker verification models.

JASA express letters·2026
Same author

Neural Responses to Affective Sentences Reveal Signatures of Depression.

Translational psychiatry·2026
Same author

Time-resolved EEG decoding reveals altered neural dynamics of affective semantic evaluation in depression and suicidality.

Communications biology·2026
Same author

Improved dynamic MRI of the wrist and heart at 0.55 T enabled by rapid 3D printed flexible coils.

Nature communications·2026
Same author

Dynamic Mode Decomposition (DMD) for Low-Latency Real-Time Cardiac MRI.

Magnetic resonance in medicine·2026

Related Experiment Video

Updated: Dec 13, 2025

Neuroimaging-Guided TMS–EEG for Real-Time Cortical Network Mapping
09:55

Neuroimaging-Guided TMS–EEG for Real-Time Cortical Network Mapping

Published on: June 13, 2025

2.1K

Deblurring for spiral real-time MRI using convolutional neural networks.

Yongwan Lim1, Yannick Bliesener1, Shrikanth Narayanan1

  • 1Ming Hsieh Department of Electrical and Computer Engineering, Viterbi School of Engineering, University of Southern California, Los Angeles, California, USA.

Magnetic Resonance in Medicine
|July 26, 2020
PubMed
Summary

This study introduces a fast convolutional neural network method for deblurring real-time MRI (RT-MRI) used in speech production. This technique significantly improves image quality and scan efficiency for dynamic imaging applications.

Keywords:
artifact correctionconvolutional neural networksdeblurringoff-resonancereal-time MRIspeech production

More Related Videos

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging
17:06

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging

Published on: November 8, 2012

26.8K
High-resolution Functional Magnetic Resonance Imaging Methods for Human Midbrain
10:06

High-resolution Functional Magnetic Resonance Imaging Methods for Human Midbrain

Published on: May 10, 2012

13.3K

Related Experiment Videos

Last Updated: Dec 13, 2025

Neuroimaging-Guided TMS–EEG for Real-Time Cortical Network Mapping
09:55

Neuroimaging-Guided TMS–EEG for Real-Time Cortical Network Mapping

Published on: June 13, 2025

2.1K
Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging
17:06

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging

Published on: November 8, 2012

26.8K
High-resolution Functional Magnetic Resonance Imaging Methods for Human Midbrain
10:06

High-resolution Functional Magnetic Resonance Imaging Methods for Human Midbrain

Published on: May 10, 2012

13.3K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Magnetic Resonance Imaging

Background:

  • Real-time MRI (RT-MRI) is crucial for dynamic processes like speech production.
  • Off-resonance artifacts commonly degrade the quality of spiral RT-MRI.
  • Existing deblurring methods often require field map information, limiting their applicability.

Purpose of the Study:

  • To develop and evaluate a novel, efficient convolutional neural network (CNN) based method for deblurring spiral RT-MRI.
  • To address off-resonance artifacts in speech production RT-MRI without needing field maps.
  • To enhance image quality and enable faster scan times for dynamic MRI applications.

Main Methods:

  • A 3-layer residual CNN architecture was designed, inspired by traditional deblurring techniques.
  • Spatially varying off-resonance blur was synthetically generated using discrete object approximation and field maps.
  • Extensive data augmentation from 2D human speech production RT-MRI datasets was employed.
  • The method's performance was investigated concerning off-resonance range, blur shift-invariance, and readout durations.

Main Results:

  • The CNN deblurring method demonstrated superior performance compared to a current autocalibrated method for in vivo data.
  • Performance was comparable to ideal reconstruction with perfect field map knowledge on synthetic data.
  • Enabled visualization of articulator boundaries with readouts up to 8 ms (3x longer than standard practice) at 1.5 T.
  • Achieved low-latency processing with a computation time of 12.3 ± 2.2 ms per frame.

Conclusions:

  • CNN-based deblurring offers a practical, efficient, and field map-free solution for spiral RT-MRI.
  • This approach can improve scan efficiency by 1.7-fold for speech production imaging.
  • Facilitates the use of spiral readouts at higher field strengths (e.g., 3 T), expanding imaging capabilities.