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

Defining the artificial intelligence knowledge gap in surgery: experience and perspectives from surgical resident and postgraduate.

Updates in surgery·2026
Same author

Incidence, Clinical Characteristics, Treatment and Outcomes of Intracardiac Cement Embolism After Vertebral Augmentation: A Systematic Review.

Global spine journal·2026
Same author

HCoVDB: a comprehensive database encompassing viral genomes, drug targets, and therapeutics of human coronaviruses.

Database : the journal of biological databases and curation·2025
Same author

Contrastive hypergraph collaborative filtering for transfer RNA-disease association prediction.

Briefings in bioinformatics·2025
Same author

Enhanced digital pathology image recognition via multi-attention mechanisms: the MACC-Net approach.

Scientific reports·2025
Same author

Enhanced boundary-directed lightweight approach for digital pathological image analysis in critical oncological diagnostics.

Journal of X-ray science and technology·2025

Related Experiment Video

Updated: Sep 8, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

491

Rethinking U-Net from an Attention Perspective with Transformers for Osteosarcoma MRI Image Segmentation.

Tianxiang Ouyang1, Shun Yang2, Fangfang Gou1

  • 1School of Computer Science and Engineering, Central South University, Changsha 410083, China.

Computational Intelligence and Neuroscience
|June 16, 2022
PubMed
Summary

This study introduces UATransNet, a novel AI model for segmenting osteosarcoma MRI images. UATransNet improves diagnostic accuracy, aiding early detection in resource-limited settings.

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

2.9K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K

Related Experiment Videos

Last Updated: Sep 8, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

491
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

2.9K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Osteosarcoma is a primary bone cancer prevalent in pediatric and adolescent populations.
  • Variability in MRI image morphology poses diagnostic challenges, especially in resource-limited areas.
  • Existing segmentation models struggle with accuracy and resource demands, hindering precision medicine approaches.

Purpose of the Study:

  • To develop a lightweight and accurate AI model for osteosarcoma MRI image segmentation.
  • To enhance early diagnosis and support clinical decision-making in osteosarcoma detection.
  • To address the limitations of current segmentation techniques in terms of accuracy and computational cost.

Main Methods:

  • Proposing UATransNet, a U-Net based architecture incorporating a multilevel guided self-aware attention module (MGAM).
  • Integrating transformer self-attention (TSAC) and global context aggregation (GCAC) for feature integration and context aggregation.
  • Employing dense residual learning and multiscale jump connections for improved feature extraction.

Main Results:

  • UATransNet demonstrated superior segmentation performance on over 80,000 osteosarcoma MRI images.
  • Achieved high Intersection over Union (IOU) of 0.922 ± 0.03 and Dice Similarity Coefficient (DSC) of 0.921 ± 0.04.
  • Provided accurate and efficient decision support information for physicians.

Conclusions:

  • UATransNet offers an effective solution for accurate osteosarcoma MRI segmentation.
  • The model's lightweight design and high accuracy are beneficial for clinical applications, especially in areas with limited resources.
  • This AI-driven approach supports precision medicine by improving diagnostic capabilities for osteosarcoma.