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

Hypofractionated Proton Reirradiation for Recurrent Glioblastoma: Clinical and Dosimetric Outcomes from a Large Single Institution Series.

International journal of radiation oncology, biology, physics·2026
Same authorSame journal

Physics-constrained dual-domain network for CBCT reconstruction from orthogonal X-rays in gynecologic radiotherapy.

Medical physics·2026
Same author

FAST-02: Results from the second in-human prospective evaluation of single-fraction proton FLASH for symptomatic thoracic bone metastases.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026
Same author

A generalist biomedical vision-language model via multi-CLIP knowledge distillation.

Nature communications·2026
Same author

Domain-adapted foundation model for automated cardiac CT substructure segmentation for thoracic radiotherapy.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026
Same author

Transformer-based synthesis of 7 Tesla-like T1 contrast from routine clinical magnetic resonance imaging.

Medical physics·2026

Related Experiment Video

Updated: Jul 16, 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

541

Deformable lung 4DCT image registration via landmark-driven cycle network.

Luke Matkovic1, Yang Lei1, Yabo Fu2

  • 1Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.

Medical Physics
|September 14, 2023
PubMed
Summary
This summary is machine-generated.

A novel deep learning method, the landmark-driven cycle network, achieves accurate lung 4DCT deformable image registration. This automated approach improves respiratory motion quantification for better patient treatment.

Keywords:
deep learningdeformable image registrationlung CTradiotherapy

More Related Videos

Four-Dimensional CT Analysis Using Sequential 3D-3D Registration
05:05

Four-Dimensional CT Analysis Using Sequential 3D-3D Registration

Published on: November 23, 2019

8.0K
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

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

Published on: April 12, 2024

626

Related Experiment Videos

Last Updated: Jul 16, 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

541
Four-Dimensional CT Analysis Using Sequential 3D-3D Registration
05:05

Four-Dimensional CT Analysis Using Sequential 3D-3D Registration

Published on: November 23, 2019

8.0K
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

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

Published on: April 12, 2024

626

Area of Science:

  • Medical Imaging
  • Radiology
  • Artificial Intelligence

Background:

  • Accurate lung 4DCT image registration is crucial for quantifying respiratory motion and optimizing treatment management.
  • Existing methods require improvement in automation, accuracy, and efficiency.

Purpose of the Study:

  • To develop a weakly supervised deep learning method for 4DCT lung deformable image registration (DIR).
  • To introduce a novel landmark-driven cycle network for enhanced DIR.

Main Methods:

  • A generator-discriminator network performs DIR, outputting deformation vector fields (DVFs).
  • The generator is optimized bidirectionally, and landmarks provide weak supervision via landmark-driven loss.
  • A discriminator regularizes DVFs by assessing the realism of the deformed CT.

Main Results:

  • The proposed method outperformed existing deep learning approaches on the DIR-Lab dataset, achieving a mean TRE of 1.20 ± 0.72 mm.
  • Bi-directional and landmark-driven loss effectively improved registration accuracy.
  • Clinical datasets showed promising results with MAE of 32.1 ± 11.6 HU and SSIM of 0.979 ± 0.011.

Conclusions:

  • The landmark-driven cycle network is a validated and effective tool for automatic lung 4DCT deformable image registration.
  • The method demonstrates performance comparable to or exceeding current state-of-the-art techniques.