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

Deep PACBED: Multitask analysis of PACBED images using deep neural networks.

Ultramicroscopy·2026
Same author

Fair Workweek laws in the US: An appraisal of intended and unintended consequences.

Science advances·2026
Same author

Points-Based Attendance Systems Associated With Presenteeism Despite Paid Sick Leave Protections.

Health affairs (Project Hope)·2026
Same author

Task interruptions impair visuospatial working memory: Behavioral and EEG evidence for feature-specific cognitive interference.

Brain research·2026
Same author

Barriers to Care Among Elderly Patients Diagnosed With Degenerative Conditions of the Lumbar Spine.

Clinical spine surgery·2026
Same author

Clinical Outcomes and Patterns of Neurological Toxicity After Stereotactic Body Radiotherapy Reirradiation (reSBRT) of Spine Metastases Previously Treated with SBRT.

Cancers·2026

Related Experiment Video

Updated: Jul 23, 2025

Whole Body and Regional Quantification of Active Human Brown Adipose Tissue Using 18F-FDG PET/CT
10:30

Whole Body and Regional Quantification of Active Human Brown Adipose Tissue Using 18F-FDG PET/CT

Published on: April 1, 2019

8.9K

Abdominal fat quantification using convolutional networks.

Daniel Schneider1,2, Tobias Eggebrecht1,3, Anna Linder1

  • 1Department of Diagnostic and Interventional Radiology, Leipzig University Hospital, Liebigstr. 20, Haus 4, 04103, Leipzig, Germany.

European Radiology
|July 12, 2023
PubMed
Summary

Automated adipose tissue quantification using deep learning fully convolutional networks (FCN) demonstrated high accuracy and reliability in patients with obesity. This advanced method offers a promising alternative to manual analysis for body composition assessment.

Keywords:
Adipose tissueDeep learningImage processing, computer-assistedMagnetic resonance imagingObesity

More Related Videos

Author Spotlight: A Non-Invasive Tool to Assess and Differentiate Fat Patterns in Liver Using 3D Dixon MRI
05:37

Author Spotlight: A Non-Invasive Tool to Assess and Differentiate Fat Patterns in Liver Using 3D Dixon MRI

Published on: October 20, 2023

1.4K
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
13:35

Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos

Published on: March 21, 2021

10.6K

Related Experiment Videos

Last Updated: Jul 23, 2025

Whole Body and Regional Quantification of Active Human Brown Adipose Tissue Using 18F-FDG PET/CT
10:30

Whole Body and Regional Quantification of Active Human Brown Adipose Tissue Using 18F-FDG PET/CT

Published on: April 1, 2019

8.9K
Author Spotlight: A Non-Invasive Tool to Assess and Differentiate Fat Patterns in Liver Using 3D Dixon MRI
05:37

Author Spotlight: A Non-Invasive Tool to Assess and Differentiate Fat Patterns in Liver Using 3D Dixon MRI

Published on: October 20, 2023

1.4K
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
13:35

Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos

Published on: March 21, 2021

10.6K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Accurate adipose tissue quantification is crucial for assessing obesity-related health risks.
  • Current semiautomated methods for adipose tissue segmentation are time-consuming and prone to reader variability.
  • Advanced computational techniques are needed to improve efficiency and reliability in body composition analysis.

Purpose of the Study:

  • To develop and evaluate a software tool for automated adipose tissue quantification from abdominal MRI data.
  • To compare the performance of fully convolutional networks (FCN) against a semiautomated reference method.
  • To assess the accuracy, reliability, processing effort, and time efficiency of the automated approach.

Main Methods:

  • Retrospective analysis of abdominal MRI data from patients with obesity.
  • Development of UNet-based FCN architectures with data augmentation for adipose tissue segmentation.
  • Ground truth established using semiautomated region-of-interest histogram thresholding.
  • Cross-validation using similarity and error measures to evaluate performance.

Main Results:

  • FCN models achieved high Dice coefficients for subcutaneous adipose tissue (SAT) (up to 0.954) and visceral adipose tissue (VAT) (up to 0.889).
  • Volumetric SAT and VAT assessments showed excellent correlation (Pearson's r > 0.997), minimal bias (<0.8%), and low standard deviation (<3.1%).
  • High intraclass correlation coefficients (SAT: 0.999, VAT: 0.996) and low coefficients of variation (SAT: 1.4%, VAT: 3.1%) indicate strong reliability.

Conclusions:

  • The developed FCN-based software provides accurate and reliable automated adipose tissue quantification.
  • This automated approach significantly reduces processing effort and reader dependence compared to semiautomated methods.
  • Deep learning techniques, specifically FCNs, are well-suited for routine, image-based body composition analysis in clinical settings.