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

Functional MRI in PSC: translating prognostic potential into clinical practice.

European radiology·2026
Same author

From static prediction to dynamic cancer trajectories: Virtual Human Twins for breast cancer decision support.

PLOS digital health·2026
Same author

Functional Liver Imaging Score (FLIS): A Prognostic Biomarker for Acute-on-Chronic Liver Failure and Liver-Related Mortality.

JHEP reports : innovation in hepatology·2026
Same author

Correcting motion-related B<sub>0</sub> inhomogeneities in magnetic resonance imaging via combined spherical harmonics and AC/DC matrix coils using DL-based prediction-simulation study.

Scientific reports·2026
Same author

Non-contiguous computed tomography lung scans can be manipulated to permit artificial intelligence analyses for interstitial lung disease in systemic sclerosis.

Biology methods & protocols·2026
Same author

From shortage to synergy: transforming global radiology workforce capacity with digital innovation and task-shifting.

The Lancet. Oncology·2026

Related Experiment Video

Updated: Jun 30, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

395

Improving Vessel Segmentation with Multi-Task Learning and Auxiliary Data Available Only During Model Training.

Daniel Sobotka1, Alexander Herold2, Matthias Perkonigg3

  • 1Computational Imaging Research Lab, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria.

Computerized Medical Imaging and Graphics : the Official Journal of the Computerized Medical Imaging Society
|March 22, 2024
PubMed
Summary

This study introduces a multi-task learning framework for segmenting liver vessels in non-contrast MRI. Auxiliary contrast-enhanced MRI data during training improves segmentation accuracy, reducing the need for extensive annotations.

Keywords:
Fully convolutional networkImage translationLiver vessel segmentationMulti-task learning

More Related Videos

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

2.7K
Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench
11:38

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

9.8K

Related Experiment Videos

Last Updated: Jun 30, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

395
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

2.7K
Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench
11:38

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

9.8K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Computational Biology

Background:

  • Liver vessel segmentation in MRI is crucial for analyzing vascular remodeling in diffuse liver diseases.
  • Current methods often require contrast-enhanced MRI, which is not always available.
  • Segmenting vessels in non-contrast MRI is challenging and data-intensive.

Purpose of the Study:

  • To develop a multi-task learning framework for liver vessel segmentation in non-contrast MRI.
  • To leverage auxiliary contrast-enhanced MRI data during training to improve segmentation performance.
  • To reduce the dependency on large-scale annotated datasets for non-contrast MRI segmentation.

Main Methods:

  • A multi-task learning framework was designed to utilize paired native and contrast-enhanced MRI data.
  • The model was trained using data with and without vessel annotations.
  • Auxiliary contrast-enhanced data was used exclusively during the training phase.

Main Results:

  • The proposed framework significantly improved the accuracy of liver vessel segmentation in non-contrast MRI.
  • The benefits were most pronounced when limited annotated data was available for training.
  • The approach demonstrated generalizability by improving brain tumor segmentation models.

Conclusions:

  • Auxiliary contrast-enhanced MRI data can effectively augment annotations for vessel segmentation in non-contrast images.
  • Multi-task learning enhances feature representation, reducing the need for extensive expert annotations.
  • This method offers a viable solution for liver vessel segmentation when contrast-enhanced sequences are unavailable.