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

Magnetic Resonance Imaging01:24

Magnetic Resonance Imaging

5.2K
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
5.2K

You might also read

Related Articles

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

Sort by
Same author

KerSpecGen: Co-piloting formal Kernel specification synthesis with refined knowledge graphs and large language models.

PloS one·2025
Same author

UWV-Yolox: A Deep Learning Model for Underwater Video Object Detection.

Sensors (Basel, Switzerland)·2023
Same author

An Instance Segmentation Model Based on Deep Learning for Intelligent Diagnosis of Uterine Myomas in MRI.

Diagnostics (Basel, Switzerland)·2023
Same author

Dimension-controlled formation of crease patterns on soft solids.

Soft matter·2016
Same author

Modeling Day-to-day Flow Dynamics on Degradable Transport Network.

PloS one·2016
Same author

Tetramethylpyrazine Protects Against Glucocorticoid-Induced Apoptosis by Promoting Autophagy in Mesenchymal Stem Cells and Improves Bone Mass in Glucocorticoid-Induced Osteoporosis Rats.

Stem cells and development·2016

Related Experiment Video

Updated: Jul 18, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

441

WA-ResUNet: A Focused Tail Class MRI Medical Image Segmentation Algorithm.

Haixia Pan1, Bo Gao1, Wenpei Bai2

  • 1College of Software, Beihang University, Beijing 100191, China.

Bioengineering (Basel, Switzerland)
|August 26, 2023
PubMed
Summary

This study introduces a novel weighted attention ResUNet (WA-ResUNet) model to improve medical image segmentation for small, rare lesions in uterine MRI scans. The new method enhances lesion identification and overall segmentation accuracy, particularly for underrepresented classes.

Keywords:
attention mechanismclass rebalancinglong-tailed distributionmedical image segmentation

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

2.8K
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

9.9K

Related Experiment Videos

Last Updated: Jul 18, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

441
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

2.8K
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

9.9K

Area of Science:

  • Medical imaging
  • Artificial intelligence in medicine
  • Computer-aided diagnosis

Background:

  • Medical image segmentation aids lesion identification but struggles with small, rare lesions and imbalanced datasets.
  • Magnetic Resonance Imaging (MRI) grayscale images present challenges in feature extraction and distinguishing valid from invalid features.
  • Existing methods do not adequately address data uncertainty and the long-tailed distribution common in medical datasets.

Purpose of the Study:

  • To develop an improved medical image segmentation model addressing limitations in identifying small/rare lesions and handling imbalanced data.
  • To enhance feature extraction and discrimination in uterine MRI scans.
  • To improve the model's performance on low-frequency classes and overall segmentation accuracy.

Main Methods:

  • Proposed a novel weighted attention ResUNet (WA-ResUNet) model incorporating attention mechanisms.
  • Developed a class weight formula based on the number of images per class to rebalance learning efficiency.
  • Evaluated the WA-ResUNet model on a uterine MRI dataset, comparing its performance against the standard ResUNet.

Main Results:

  • The WA-ResUNet model significantly improved Intersection over Union (IoU) for the low-frequency class (Nabothian cysts) by 21.87%.
  • Overall mean Intersection over Union (mIoU) demonstrated a substantial increase of over 6.5% compared to the baseline ResUNet.
  • The model showed enhanced attention to both valid and invalid features, improving segmentation performance.

Conclusions:

  • The proposed WA-ResUNet model effectively addresses challenges in segmenting small, rare lesions and imbalanced medical imaging data.
  • The weighted attention mechanism and class weight formula enhance model performance and learning efficiency.
  • This approach offers a promising advancement for accurate lesion detection in uterine MRI and other medical imaging applications.