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

You might also read

Related Articles

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

Sort by
Same author

Evaluation of Posture-Dependent Signal Intensity and Contrast Alterations in Low-Field Brain Magnetic Resonance Imaging.

Diagnostics (Basel, Switzerland)·2026
Same author

Joint Clustering and Power Optimization for the SPMA Protocol in UAV Swarm Communication in Frequency-Constrained Scenarios.

Sensors (Basel, Switzerland)·2026
Same author

A Hybrid Vision Transformer-BiRNN Architecture for Direct k-Space to Image Reconstruction in Accelerated MRI.

Journal of imaging·2026
Same author

A review of deep learning-based reconstruction methods for accelerated MRI using spatiotemporal and multi-contrast redundancies.

Biomedical engineering letters·2024
Same author

Self-supervised learning for denoising of multidimensional MRI data.

Magnetic resonance in medicine·2024
Same author

A noise robust image reconstruction using slice aware cycle interpolator network for parallel imaging in MRI.

Medical physics·2024

Related Experiment Video

Updated: Oct 17, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.1K

Unsupervised anomaly detection in MR images using multicontrast information.

Byungjai Kim1, Kinam Kwon2, Changheun Oh1

  • 1Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Guseong-dong, Yuseong-gu, Daejeon, Republic of Korea.

Medical Physics
|October 10, 2021
PubMed
Summary

This study introduces an unsupervised deep learning algorithm for detecting anomalies in multi-contrast magnetic resonance imaging (MRI). The method effectively identifies diseases like glioblastoma and stroke lesions using normal MRI data for training.

Keywords:
anomaly detectionmagnetic resonance imagingmulticontrast imagessingularity problemunsupervised learning

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

692
Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
07:34

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

8.2K

Related Experiment Videos

Last Updated: Oct 17, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.1K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

692
Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
07:34

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

8.2K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Biomedical Engineering

Background:

  • Anomaly detection in MRI is crucial for identifying disease biomarkers.
  • Distinguishing diseased tissue from normal tissue is a key challenge in medical imaging analysis.

Purpose of the Study:

  • To propose an unsupervised algorithm for pixel-level anomaly detection in multi-contrast MRI.
  • To develop a deep learning framework that utilizes only normal MRI images for training.

Main Methods:

  • A deep neural network with feature generation and density estimation stages was developed.
  • Feature generation involved contrast translation and dimension reduction.
  • Density estimation used a Gaussian Mixture Model (GMM) to model normal data distributions.

Main Results:

  • The algorithm was successfully applied to detect glioblastoma and ischemic stroke lesions.
  • Quantitative and qualitative evaluations showed significant improvements over six existing anomaly detection methods.
  • Ablation studies confirmed the effectiveness of individual components within the proposed framework.

Conclusions:

  • The proposed deep learning framework is effective for anomaly detection in multi-contrast MRI.
  • Unsupervised approaches show promise for lesion detection where annotated data is scarce.