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

Residuals and Least-Squares Property01:11

Residuals and Least-Squares Property

7.4K
The vertical distance between the actual value of y and the estimated value of y. In other words, it measures the vertical distance between the actual data point and the predicted point on the line
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...
7.4K
Reducing Line Loss01:18

Reducing Line Loss

173
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
173

You might also read

Related Articles

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

Sort by
Same author

Programming Multidomain Peptides With Molecular Frustration Into Biomolecular Condensates.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Atomic-Level Polishing of Single-Crystal Diamond Using a Combination of Reactive Ion Etching and Chemical Mechanical Polishing.

Materials (Basel, Switzerland)·2026
Same author

Cardiac-Derived ECM Microspheres for Enhanced hiPSC-CMs Maturation.

Advanced functional materials·2026
Same author

Functional Design of Near-Infrared Aggregation-Induced Emission Photosensitizers Targeting Cell Membrane and Mitochondria Enhance Photodynamic Efficiency.

Advanced healthcare materials·2026
Same author

Intrapericardially injected hydrogel loaded with stromal cell secretome microparticles improves post-infarction myocardial repair in pigs.

European heart journal·2026
Same author

A simplified prediction model for in-hospital mortality in children with congenital heart disease after surgery.

Translational pediatrics·2026

Related Experiment Video

Updated: Jul 18, 2025

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

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

Published on: October 27, 2023

1.2K

R2Net: Efficient and flexible diffeomorphic image registration using Lipschitz continuous residual networks.

Ankita Joshi1, Yi Hong2

  • 1School of Computing, University of Georgia, Athens, 30602, USA.

Medical Image Analysis
|August 20, 2023
PubMed
Summary

This study introduces Residual Registration Network (R2Net), a deep learning framework for fast and accurate 3D image registration. R2Net handles large deformations efficiently, offering comparable or better accuracy than traditional methods.

Keywords:
Deep residual networksLipschitz continuityMulti-scale registrationStationary and non-stationary velocity fieldsUnsupervised diffeomorphic image registration

More Related Videos

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

48.1K
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

884

Related Experiment Videos

Last Updated: Jul 18, 2025

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

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

Published on: October 27, 2023

1.2K
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

48.1K
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

884

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Machine Learning

Background:

  • Classical diffeomorphic image registration is accurate but computationally expensive.
  • Existing deep learning methods often sacrifice diffeomorphism or struggle with large deformations, limited by stationary velocity fields (SVFs) and inefficient integration techniques.
  • Handling large 3D image volumes remains a challenge for deep learning registration.

Purpose of the Study:

  • To develop an unsupervised deep learning framework for accurate and efficient diffeomorphic image registration.
  • To enable the capture of large deformations while reducing computational costs.
  • To improve the handling of large 3D image volumes in medical image registration.

Main Methods:

  • Introduced Residual Registration Network (R2Net), utilizing deep residual networks (ResNets) to approximate continuous diffeomorphic transformations.
  • Employed flexible parameterization with stationary or time-varying velocity fields for enhanced deformation capture and reduced integration costs.
  • Incorporated a Lipschitz continuity constraint for guaranteed diffeomorphic deformations and a hierarchical, multi-phase learning strategy for large volumes.

Main Results:

  • R2Net achieved comparable or superior registration accuracy to classical methods (SyN, diffeomorphic VoxelMorph) across diverse 3D datasets (brain MRI, cardiac MRI, lung CT).
  • The framework demonstrated significantly reduced time and memory costs compared to existing deep learning approaches.
  • Generated deformations were smoother, indicating improved diffeomorphic properties.

Conclusions:

  • R2Net offers an effective unsupervised deep learning solution for diffeomorphic image registration, balancing accuracy, speed, and flexibility.
  • The proposed methods address key limitations of prior deep learning registration techniques, particularly for large deformations and volumes.
  • The framework shows promise for various 3D medical imaging applications requiring precise and efficient registration.