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

Spatiotemporal Gaussian representation-based dynamic reconstruction and motion estimation framework for time-resolved volumetric MR imaging (DREME-GSMR).

ArXiv·2026
Same author

'See' through the surface: surface-derived three-dimensional AI-driven real-time imaging solution for intra-treatment image guidance.

Machine Learning. Health·2026
Same author

FDDM: Unsupervised Medical Image Translation with a Frequency-Decoupled Diffusion Model.

Machine learning: science and technology·2026
Same author

TSMS-SAM2: Multi-scale Temporal Sampling Augmentation and Memory-Splitting Pruning for Promptable Video Object Segmentation and Tracking in Surgical Scenarios.

Machine Learning. Health·2026
Same author

cMeta-INR: cohort-informed meta-learning-based implicit neural representation for deformable registration-driven real-time volumetric MRI estimation.

Physics in medicine and biology·2025
Same author

Prior-adapted progressive time-resolved CBCT reconstruction using a dynamic reconstruction and motion estimation method.

Medical physics·2025

Related Experiment Video

Updated: Oct 17, 2025

Three-dimensional Location Approach with Silk Thread Guided Laparoscopic Segmentectomy for Liver Tumor
06:39

Three-dimensional Location Approach with Silk Thread Guided Laparoscopic Segmentectomy for Liver Tumor

Published on: May 23, 2025

180

Automatic liver tumor localization using deep learning-based liver boundary motion estimation and biomechanical

Hua-Chieh Shao1, Xiaokun Huang1, Michael R Folkert1

  • 1Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA.

Medical Physics
|October 11, 2021
PubMed
Summary

A deep-learning (DL) method improved liver tumor localization by optimizing deformation-vector-fields (DVFs) at the liver boundary. This enhances biomechanical modeling accuracy for better on-board tumor tracking during radiation therapy.

Keywords:
CBCTbiomechanical modelingconvolutional neural networkdeep learningdeformable registrationliver

More Related Videos

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.4K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.0K

Related Experiment Videos

Last Updated: Oct 17, 2025

Three-dimensional Location Approach with Silk Thread Guided Laparoscopic Segmentectomy for Liver Tumor
06:39

Three-dimensional Location Approach with Silk Thread Guided Laparoscopic Segmentectomy for Liver Tumor

Published on: May 23, 2025

180
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.4K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.0K

Area of Science:

  • Medical Imaging
  • Computational Biology
  • Radiotherapy

Background:

  • Two-dimensional-to-three-dimensional (2D-3D) deformable registration aids liver tumor localization.
  • Biomechanical modeling refines 2D-3D registration using tissue elasticity.
  • Low contrast at the caudal liver boundary hinders registration accuracy and biomechanical modeling.

Purpose of the Study:

  • Develop a deep-learning (DL) method to optimize liver boundary deformation-vector-fields (DVFs).
  • Improve accuracy of biomechanical modeling for intra-liver tumor localization.
  • Enhance on-board tumor localization using cone-beam computed tomography (CBCT).

Main Methods:

  • A U-Net based DL network was trained to correlate cranial and caudal liver boundary DVFs.
  • Network inputs were 3D DVFs from 2D-3D registration, masked by liver boundary structures.
  • Optimized DVFs were used for biomechanical modeling to refine tumor motion solutions.

Main Results:

  • The DL method improved DVFs accuracy at the liver boundary.
  • Average tumor localization errors (center-of-mass-errors) were reduced: 4.7mm (2D-3D), 2.9mm (2D-3D-Bio), 1.7mm (DL-Bio).
  • DICE coefficients improved from 0.60 (2D-3D) to 0.78 (DL-Bio).

Conclusions:

  • DL-Bio optimizes DVFs at the liver boundary, improving biomechanical modeling.
  • Enhanced boundary conditions lead to more accurate intra-liver low-contrast tumor localization.
  • This approach advances automated tumor localization in radiotherapy.