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

Computed Tomography01:10

Computed Tomography

7.9K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
7.9K

You might also read

Related Articles

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

Sort by
Same author

Integrating microbial biomass, composition and function to discern the level of anthropogenic activity in a river ecosystem.

Environment international·2018
Same author

c-Jun-mediated microRNA-302d-3p induces RPE dedifferentiation by targeting p21<sup>Waf1/Cip1</sup>.

Cell death & disease·2018
Same author

High expression of synthesis of cytochrome c oxidase 2 and TP53-induced glycolysis and apoptosis regulator can predict poor prognosis in human lung adenocarcinoma.

Human pathology·2018
Same author

Comparison of the efficacy of dispensing granules with traditional decoction: a systematic review and meta-analysis.

Annals of translational medicine·2018
Same author

Serum Wisteria floribunda agglutinin-positive Mac-2-binding protein evaluates liver function and predicts prognosis in liver cirrhosis.

Journal of digestive diseases·2018
Same author

Acupuncture for constipation in patients with stroke: protocol of a systematic review and meta-analysis.

BMJ open·2018

Related Experiment Video

Updated: Jan 10, 2026

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

8.4K

PF-DAformer: Proximal Femur Segmentation via Domain Adaptive Transformer for Dual-Center QCT.

Rochak Dhakal1, Chen Zhao2, Zixin Shi1

  • 1Department of Applied Computing, Michigan Technological University, Houghton, MI, 49931, USA.

Arxiv
|November 24, 2025
PubMed
Summary

This study introduces a domain-adaptive transformer for segmenting proximal femur bone density using quantitative computed tomography (QCT). The method overcomes domain shift across institutions, ensuring reliable bone strength and fracture risk assessments for multi-center research.

Keywords:
domain adaptationhip fractureimage segmentationquantitative computed tomographytransformers

More Related Videos

Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin
09:36

Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin

Published on: March 14, 2018

9.7K
In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy
07:43

In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy

Published on: July 2, 2021

3.5K

Related Experiment Videos

Last Updated: Jan 10, 2026

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

8.4K
Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin
09:36

Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin

Published on: March 14, 2018

9.7K
In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy
07:43

In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy

Published on: July 2, 2021

3.5K

Area of Science:

  • Medical Imaging
  • Biomedical Engineering
  • Artificial Intelligence

Background:

  • Quantitative computed tomography (QCT) is vital for assessing bone strength and fracture risk via proximal femur density analysis.
  • Automated segmentation models struggle with domain shift across institutions, hindering multi-center research and reproducibility.
  • Variations in scanners, settings, and demographics cause unstable predictions and unreliable quantitative metrics.

Purpose of the Study:

  • To develop a domain-adaptive transformer segmentation framework for multi-institutional QCT analysis.
  • To address domain shift challenges in proximal femur segmentation for improved bone density assessment.
  • To ensure reproducible radiomics and finite element analysis results across different clinical sites.

Main Methods:

  • Developed a 3D TransUNet backbone incorporating adversarial alignment (Gradient Reversal Layer - GRL) and statistical alignment (Maximum Mean Discrepancy - MMD).
  • Trained and validated the model on a large hip fracture cohort (1,024 scans from Tulane, 384 from Rochester).
  • Integrated GRL to discourage site-specific encoding and MMD to reduce distributional mismatches between institutions.

Main Results:

  • Achieved high segmentation accuracy: Dice 99.53%, Precision 99.64%, Hausdorff Distance 0.77 mm, significantly outperforming non-adaptive baselines (p < 0.01).
  • Demonstrated preservation of radiomic features, with Pearson correlation coefficients > 0.99 compared to ground truth.
  • The combined GRL and MMD domain adaptation strategy yielded the most consistent and accurate performance.

Conclusions:

  • The proposed domain-adaptive transformer framework effectively overcomes domain shift in multi-institutional QCT segmentation.
  • This approach ensures scanner-agnostic feature learning while maintaining anatomical detail and quantitative accuracy.
  • The method enhances reproducibility for multi-center osteoporosis research, radiomics, and finite element analysis.