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

Robust Construction of Diffusion MRI Atlases with Correction for Inter-Subject Fiber Dispersion.

Computational diffusion MRI : MICCAI Workshop·2017
Same author

Robust Fusion of Diffusion MRI Data for Template Construction.

Scientific reports·2017
Same author

Learning-Based Multimodal Image Registration for Prostate Cancer Radiation Therapy.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2017
Same author

Segmenting hippocampal subfields from 3T MRI with multi-modality images.

Medical image analysis·2017
Same author

Joint Discriminative and Representative Feature Selection for Alzheimer's Disease Diagnosis.

Machine learning in medical imaging. MLMI (Workshop)·2017
Same author

Single- and Multiple-Shell Uniform Sampling Schemes for Diffusion MRI Using Spherical Codes.

IEEE transactions on medical imaging·2017

Related Experiment Video

Updated: Aug 7, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.2K

Deep learning based brain MRI registration driven by local-signed-distance fields of segmentation maps.

Yue Yang1, Shunbo Hu1, Lintao Zhang1

  • 1School of Information Science and Engineering, Linyi University, Linyi, Shandong, China.

Medical Physics
|March 7, 2023
PubMed
Summary

This study introduces a dually-supervised registration method using local-signed-distance fields (LSDFs) to improve brain MRI registration accuracy and plausibility. The novel approach enhances correspondence by incorporating geometric information alongside intensity and segmentation data.

Keywords:
deep learninglocal-signed-distance fieldsmedical image registration

More Related Videos

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K
Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
12:50

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly

Published on: April 14, 2014

40.3K

Related Experiment Videos

Last Updated: Aug 7, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.2K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K
Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
12:50

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly

Published on: April 14, 2014

40.3K

Area of Science:

  • Medical Image Analysis
  • Deep Learning
  • Neuroimaging

Background:

  • Unsupervised deep learning registration relies on image intensity, which can be affected by variations.
  • Combining unsupervised and weakly-supervised methods (dually-supervised registration) aims to improve accuracy.
  • Directly using segmentation labels in registration can lead to implausible deformations, particularly at tissue edges.

Purpose of the Study:

  • To enhance brain MRI registration accuracy and plausibility by introducing local-signed-distance fields (LSDFs).
  • To dually supervise registration using both image intensity and voxelwise geometric distance information.
  • To ensure accurate voxelwise correspondence both inside and outside tissue edges.

Main Methods:

  • Construct LSDFs from segmentation labels to provide enhanced geometric guidance.
  • Develop an LSDF-Net utilizing 3D dilation and erosion layers for LSDF calculation.
  • Integrate VoxelMorph (VM) and LSDF-Net into a dually-supervised network (VMLSDF) for combined intensity and geometric supervision.

Main Results:

  • VMLSDF demonstrated superior performance over unsupervised VM and VMseg on four public datasets (LPBA40, HBN, OASIS1, OASIS3).
  • Improved Dice Similarity Coefficient (DSC) and 95% Hausdorff Distance (HD) were observed with VMLSDF.
  • Reduced percentage of negative Jacobian determinant (NJD) indicates more plausible deformations with VMLSDF compared to VMseg.

Conclusions:

  • LSDFs significantly improve registration accuracy compared to existing VM and VMseg methods.
  • The proposed method enhances the plausibility of dense deformation fields (DDFs) over VMseg.
  • Dually-supervised registration incorporating LSDFs offers a more robust and accurate approach for brain MRI alignment.