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 Experiment Video

Updated: Jun 17, 2026

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

Implementation and evaluation of various demons deformable image registration algorithms on a GPU.

Xuejun Gu1, Hubert Pan, Yun Liang

  • 1Department of Radiation Oncology, University of California San Diego, La Jolla, CA 92037, USA.

Physics in Medicine and Biology
|December 17, 2009
PubMed
Summary

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

Organ preservation in rectal cancer following clinical complete response after short-course radiotherapy-based total neoadjuvant therapy.

Clinical and translational radiation oncology·2026
Same author

Quantitative perfusion imaging from non-contrast micro-ct for pulmonary embolism evaluation in preclinical models.

Physics in medicine and biology·2026
Same author

Attentive pre-training question embeddings for knowledge tracing with semantically-enhanced knowledge structure and concept label-guided heterogeneous graph representation.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

From models to agentic systems: the evolution of artificial intelligence in medical physics.

Physics in medicine and biology·2026
Same author

Base excision repair hierarchy in eukaryotes: Intrinsically disordered region-mediated regulation of genomic surveillance and assembly dynamics.

Current opinion in structural biology·2026
Same author

Diagnoses of pulmonary embolism from non-contrast 4DCT using image processing-derived quantitative perfusion scores.

npj biomedical innovations·2026

Accelerating radiation therapy with graphics processing units (GPUs) enables faster image segmentation. The original demons algorithm on GPUs achieved accurate and efficient deformable image registration for online adaptive radiation therapy.

Area of Science:

  • Medical Physics
  • Radiotherapy Technology
  • Medical Imaging

Background:

  • Online adaptive radiation therapy (ART) requires rapid image segmentation for clinical implementation.
  • Deformable image registration (DIR) is a key automated segmentation method for ART.
  • Current DIR computational times are too slow for online ART.

Purpose of the Study:

  • To accelerate deformable image registration (DIR) for online adaptive radiation therapy (ART) using graphics processing units (GPUs).
  • To evaluate the accuracy and efficiency of GPU-accelerated DIR algorithms on pulmonary 4D CT images.

Main Methods:

  • Implemented a gray-scale-based DIR algorithm (demons) and five variants on GPUs using CUDA.
  • Evaluated spatial accuracy using over 1100 expert-determined landmark pairs on pulmonary 4D CT images.

More Related Videos

Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
05:49

Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images

Published on: February 23, 2024

Related Experiment Videos

Last Updated: Jun 17, 2026

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

Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
05:49

Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images

Published on: February 23, 2024

  • Measured computational time and 3D error for GPU-based DIR.

    • Main Results:

    • GPU-based DIR achieved computation times of 7 to 11 seconds.
    • Average 3D error ranged from 1.5 to 1.8 mm across testing scenarios.
    • The original passive force demons algorithm demonstrated superior performance in accuracy, efficiency, and implementation.

    Conclusions:

    • GPU acceleration significantly reduces DIR computation time, making it viable for online ART.
    • The original demons algorithm is a highly effective and efficient choice for GPU-based DIR in ART.
    • This work addresses a critical technical barrier for the clinical adoption of online ART.