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

Comprehensive evaluation of a deep learning-based synthetic CT model for MR-only radiotherapy across multiple anatomical sites.

Physics and imaging in radiation oncology·2026
Same author

Toward universal dose prediction: A multi-scale, multi-objective framework for sequential boost radiotherapy.

Medical physics·2026
Same author

Strategies for enhancing delivery efficiency on MR-Linac: A dosimetric study and historical plan review.

Journal of applied clinical medical physics·2026
Same author

RAPID-LC: rapid evidence-to-practice uptake of large core thrombectomy across a stroke consortium.

Journal of neurology·2026
Same author

Reply: Why fan therapy does not yet demonstrate enhanced training benefits in people with chronic respiratory disease.

The European respiratory journal·2026
Same author

A Simulation-Free Radiation Therapy Workflow Using Synthetic Computed Tomography Generated from Diagnostic Magnetic Resonance Imaging for Personalized Hippocampal-Sparing Whole-Brain Treatment.

Practical radiation oncology·2026
Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026
Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026
Same journal

A novel optical respiratory gating system with a hybrid phase-amplitude algorithm for spot-scanning proton therapy.

Medical physics·2026
Same journal

Gamma Knife treatment planning using knowledge-based reinforcement learning.

Medical physics·2026
Same journal

Development and characterization of a novel, small animal external beam irradiator using a clinical high dose rate brachytherapy source.

Medical physics·2026
Same journal

Deep learning-based dose prediction for MR-guided prostate SIB: Supporting rapid feasibility assessment and adaptive editing margin selection.

Medical physics·2026
See all related articles

Related Experiment Video

Updated: Aug 23, 2025

Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform
07:57

Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform

Published on: March 24, 2022

2.9K

Segmentation by test-time optimization for CBCT-based adaptive radiation therapy.

Xiao Liang1, Jaehee Chun2, Howard Morgan1

  • 1Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA.

Medical Physics
|October 31, 2022
PubMed
Summary
This summary is machine-generated.

Test-time optimization (TTO) refines deep learning deformable image registration models for online adaptive radiotherapy, improving segmentation accuracy, especially for outlier patients. This method enhances contouring in cone-beam CT images, crucial for precise cancer treatment delivery.

Keywords:
CBCTdeep learningdeformable image registrationsegmentationtest-time optimization

More Related Videos

Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
08:25

Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System

Published on: April 11, 2018

15.4K
Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy
08:17

Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy

Published on: June 7, 2015

15.8K

Related Experiment Videos

Last Updated: Aug 23, 2025

Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform
07:57

Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform

Published on: March 24, 2022

2.9K
Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
08:25

Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System

Published on: April 11, 2018

15.4K
Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy
08:17

Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy

Published on: June 7, 2015

15.8K

Area of Science:

  • Medical Physics
  • Radiotherapy Technology
  • Medical Imaging

Background:

  • Online adaptive radiotherapy (ART) demands precise auto-segmentation of target volumes and organs-at-risk (OARs) on cone-beam computed tomography (CBCT) images.
  • CBCT images present challenges due to artifacts and poor soft-tissue contrast, hindering direct segmentation.
  • Deep learning (DL)-based deformable image registration (DIR) offers improved segmentation by propagating contours from planning CT, but population-based models face generalizability issues.

Purpose of the Study:

  • To introduce and evaluate a test-time optimization (TTO) method for refining pretrained DL-based DIR population models for online ART.
  • To enhance segmentation accuracy by adapting DIR models to individual patients and progressively to each treatment fraction.
  • To mitigate the generalizability problem inherent in population-based DL models, particularly for outlier cases.

Main Methods:

  • Proposed a test-time optimization (TTO) approach to refine a pretrained DL-based DIR population model.
  • Applied TTO to individualize the population model for each test patient and subsequently for each treatment fraction.
  • Utilized data from 239 head-and-neck squamous cell carcinoma patients, training a population model on 200 and refining it for the remaining 39 test patients.

Main Results:

  • Individualized models achieved average improvements of 0.04 in Dice Similarity Coefficient (DSC) and 0.98 mm in 95% Hausdorff Distance (HD95) across 17 structures for 39 patients.
  • Significant improvements were observed for outlier patients with substantial anatomical changes.
  • TTO-derived individualized models were generated in approximately 1 minute, with fractional adaptations taking less than a minute.

Conclusions:

  • The TTO method effectively refines DL-based DIR models for online ART, boosting segmentation accuracy.
  • TTO is particularly beneficial for outlier patients where standard population models may fail.
  • The method offers a practical solution for improving image segmentation in adaptive radiotherapy workflows.