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

Reconstruction of Signal using Interpolation01:10

Reconstruction of Signal using Interpolation

383
Signal processing techniques are essential for accurately converting continuous signals to digital formats and vice versa. When a continuous signal is sampled with a period T, the resulting sampled signal exhibits replicas of the original spectrum in the frequency domain, spaced at intervals equal to the sampling frequency. To handle this sampled signal, a zero-order hold method can be applied, which creates a piecewise constant signal by retaining each sample's value until the next...
383
Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

146
Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
146
Upsampling01:22

Upsampling

337
Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...
337
Aliasing01:18

Aliasing

263
Accurate signal sampling and reconstruction are crucial in various signal-processing applications. A time-domain signal's spectrum can be revealed using its Fourier transform. When this signal is sampled at a specific frequency, it results in multiple scaled replicas of the original spectrum in the frequency domain. The spacing of these replicas is determined by the sampling frequency.
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
263
Downsampling01:20

Downsampling

285
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
285
Linear Approximation in Time Domain01:21

Linear Approximation in Time Domain

132
Nonlinear systems often require sophisticated approaches for accurate modeling and analysis, with state-space representation being particularly effective. This method is especially useful for systems where variables and parameters vary with time or operating conditions, such as in a simple pendulum or a translational mechanical system with nonlinear springs.
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length,...
132

You might also read

Related Articles

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

Sort by
Same author

Arc-ZTE: Continuously-Slewed Zero-TE Imaging With Incoherent Temporal Sampling for Near-Silent Dynamic MRI.

Magnetic resonance in medicine·2026
Same author

Semi-supervision for clinical contrast-weighted image synthesis from magnetic resonance fingerprinting.

Magma (New York, N.Y.)·2026
Same author

AI Advancements in Healthcare: Medical Imaging and Sensing Technologies.

Bioengineering (Basel, Switzerland)·2025
Same author

DM-Net: a physics-model-independent direct mapping approach for calibration-free multi-coil MRI.

Research square·2025
Same author

Data- and Physics-Driven Deep Learning Based Reconstruction for Fast MRI: Fundamentals and Methodologies.

IEEE reviews in biomedical engineering·2024
Same author

Spherical echo-planar time-resolved imaging (sEPTI) for rapid 3D quantitative <math><mrow><msubsup><mi>T</mi> <mn>2</mn> <mo>*</mo></msubsup></mrow></math> and susceptibility imaging.

Magnetic resonance in medicine·2024

Related Experiment Video

Updated: Sep 27, 2025

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
10:44

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging

Published on: June 21, 2024

644

Improving high frequency image features of deep learning reconstructions via k-space refinement with null-space

Kanghyun Ryu1, Cagan Alkan2, Shreyas S Vasanawala1

  • 1Department of Radiology, Stanford University, 300 Pasteur Drive, Stanford, California, 94305, USA.

Magnetic Resonance in Medicine
|April 15, 2022
PubMed
Summary

This study introduces a novel null-space kernel method to refine deep learning (DL) MRI reconstructions, enhancing image details and textures lost in standard DL approaches. The technique improves high-frequency details without increasing overall image error.

Keywords:
accelerated MRIdeep learningparallel imagingunrolled neural network

More Related Videos

Author Spotlight: Using Hyperpolarized Xenon-129 MRI to Study Lung Diseases
09:55

Author Spotlight: Using Hyperpolarized Xenon-129 MRI to Study Lung Diseases

Published on: January 5, 2024

1.4K
Real-time Monitoring of High Intensity Focused Ultrasound HIFU Ablation of In Vitro Canine Livers Using Harmonic Motion Imaging for Focused Ultrasound HMIFU
07:38

Real-time Monitoring of High Intensity Focused Ultrasound HIFU Ablation of In Vitro Canine Livers Using Harmonic Motion Imaging for Focused Ultrasound HMIFU

Published on: November 3, 2015

10.1K

Related Experiment Videos

Last Updated: Sep 27, 2025

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
10:44

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging

Published on: June 21, 2024

644
Author Spotlight: Using Hyperpolarized Xenon-129 MRI to Study Lung Diseases
09:55

Author Spotlight: Using Hyperpolarized Xenon-129 MRI to Study Lung Diseases

Published on: January 5, 2024

1.4K
Real-time Monitoring of High Intensity Focused Ultrasound HIFU Ablation of In Vitro Canine Livers Using Harmonic Motion Imaging for Focused Ultrasound HMIFU
07:38

Real-time Monitoring of High Intensity Focused Ultrasound HIFU Ablation of In Vitro Canine Livers Using Harmonic Motion Imaging for Focused Ultrasound HMIFU

Published on: November 3, 2015

10.1K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Image Reconstruction

Background:

  • Deep learning (DL) accelerates Magnetic Resonance Imaging (MRI) but often loses high-frequency details.
  • Unrolled neural networks are a key DL approach for MRI reconstruction.

Purpose of the Study:

  • To propose a novel refinement method using a null-space kernel to improve blurred image details and textures in DL-based MRI reconstruction.
  • To enhance k-space data and recover high-frequency information lost during DL reconstruction.

Main Methods:

  • A null-space kernel refinement method was developed to constrain DL output to auto-calibration line relationships.
  • The method was tested on DL reconstructions across diverse datasets (fastMRI knee and brain), networks, and under-sampling schemes.

Main Results:

  • The refinement method reduced k-space structural errors and improved intensity homogeneity.
  • Reconstructed images exhibited sharper details and textures, with significant improvements in high-frequency metrics (SSIM, GMSD).
  • Overall image error (PSNR) was not compromised.

Conclusions:

  • Refining DL MRI reconstruction with the proposed null-space kernel method offers general improvements.
  • The approach enhances image quality across various datasets and network architectures.