Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Positive-Unlabeled Learning in Implicit Feedback from Data Missing-Not-At-Random Perspective.

Entropy (Basel, Switzerland)·2026

Same author

Dose Prediction in Proton Cancer Therapy based on Density Maps from Dual-energy CT Using Joint Statistical Image Reconstruction.

Proceedings of SPIE--the International Society for Optical Engineering·2025

Same author

Accurate Proton Stopping Power Images Reconstructed using Joint Statistical Dual Energy CT: Experimental Verification and Impact of Fan-Beam CT Scatter.

Proceedings of SPIE--the International Society for Optical Engineering·2025

Same author

Development of a Prototype AWSM-PET Device for Augmented Whole-Body PET Imaging and Initial Human Study.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine·2025

Same author

Derivation of tissue properties from basis-vector model weights for dual-energy CT-based Monte Carlo proton beam dose calculations.

Biomedical physics & engineering express·2025

Same author

Derivation of Tissue Properties from Basis-Vector Model Weights for Dual-Energy CT-Based Monte Carlo Proton Beam Dose Calculations.

ArXiv·2025

Search research articles

Related Experiment Video

Updated: Jun 4, 2025

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

IConDiffNet: an unsupervised inverse-consistent diffeomorphic network for medical image registration.

Rui Liao¹, Jeffrey F Williamson¹, Tianyu Xia²

¹Washington University in St. Louis, Saint Louis, MO 63130, United States of America.

Physics in Medicine and Biology

|January 2, 2025

Summary

This summary is machine-generated.

This study introduces IConDiffNet, a novel deep learning model for fast and accurate medical image registration. It ensures diffeomorphic and inverse-consistent transformations, outperforming existing methods on brain MRI datasets.

Keywords:

deep learning deformable image registration diffeomorphic inverse-consistent medical imaging

More Related Videos

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

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

Published on: September 25, 2019

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

Related Experiment Videos

Last Updated: Jun 4, 2025

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

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

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

Published on: September 25, 2019

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

Area of Science:

Medical Imaging
Computational Anatomy
Deep Learning

Background:

Deformable image registration (DIR) is vital in medical imaging.
Traditional DIR methods are computationally intensive and struggle with complex deformations.
Existing deep learning DIR models often fail to enforce diffeomorphic and inverse-consistent transformations.

Purpose of the Study:

To develop a novel unsupervised neural network for fast, accurate, and inverse-consistent diffeomorphic DIR.
To address limitations of current deep learning approaches in enforcing essential transformation properties.

Main Methods:

Introduced IConDiffNet, an unsupervised inverse-consistent diffeomorphic registration network.
Employed a novel energy constraint to minimize deformation energy.
Utilized symmetric paths with cascaded updating blocks to estimate time-dependent velocity fields for forward and inverse transformations.

Main Results:

IConDiffNet achieved fast and accurate DIR on a 3D inter-patient brain MRI dataset.
Demonstrated superior performance over state-of-the-art methods in Dice Similarity Coefficient (DSC) and Hausdorff distance.
Visualizations confirmed IConDiffNet's ability to handle complex deformations and align structures effectively.

Conclusions:

IConDiffNet advances unsupervised deep learning for DIR by ensuring inverse consistency and diffeomorphic properties.
Offers improved registration accuracy crucial for clinical applications.
The network's generalizable structure allows adaptation to various 3D image registration challenges.