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

Identification and functional validation of SPON2 as a novel biomarker for the diagnosis and prognosis in hepatocellular carcinoma.

Journal of gastrointestinal oncology·2026
Same author

Identification and prognostic analysis of genes related to CTNNB1 mutations in hepatocellular carcinoma.

Biology direct·2026
Same author

Risk factors for post-ERCP pancreatitis and assessment of stent exchange intervals in children with chronic pancreatitis: a single-center retrospective study.

BMC pediatrics·2026
Same author

Rsk4 Mediates Brain FGF19 Signaling to Constrain Diet-Induced Obesity in Mice.

Diabetes·2026
Same author

Subtype-Specific Prognosis, Recurrence Patterns, and Molecular Features in 148 Chinese Uterine Sarcomas: A Real-World Study.

Cancers·2026
Same author

River Surface Velocity and Discharge Estimation Using Optical Flow and Unlabeled Physics-Informed Neural Networks.

Sensors (Basel, Switzerland)·2026

Related Experiment Video

Updated: Aug 29, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

485

COVID-19 CT image segmentation method based on swin transformer.

Weiwei Sun1, Jungang Chen1, Li Yan2

  • 1Chongqing University of Posts and Telecommunication, Chongqing, China.

Frontiers in Physiology
|September 8, 2022
PubMed
Summary
This summary is machine-generated.

This study introduces an improved U-Net deep learning model for segmenting COVID-19 lung lesions in CT scans. The enhanced model aids doctors in diagnosis and analysis, improving efficiency and reducing infection risk.

Keywords:
COVID-19CT imagedeep learningdetection and recognitionlesion segmentation

More Related Videos

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K
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

Related Experiment Videos

Last Updated: Aug 29, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

485
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K
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

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Pulmonology

Background:

  • The COVID-19 pandemic necessitates efficient diagnostic tools for lung lesion analysis.
  • Computed Tomography (CT) is crucial for visualizing COVID-19 related lung abnormalities.
  • Deep learning offers potential for automated screening and quantitative assessment.

Purpose of the Study:

  • To develop an advanced deep learning model for accurate segmentation of COVID-19 lung lesions in CT images.
  • To enhance diagnostic efficiency and reduce infection risk for healthcare professionals.
  • To improve the quantitative analysis of lung lesions in COVID-19 patients.

Main Methods:

  • A modified U-Net architecture incorporating atrous convolution and a convolutional block attention module (CBAM).
  • Integration of Swin Transformer in the encoder for parameter reduction and performance enhancement.
  • Training and validation using the CC-CCII lesion segmentation dataset comprising 750 annotated CT images.

Main Results:

  • The proposed model achieved a mean pixel accuracy of 87.62%, mean intersection over union of 80.6%, and Dice similarity coefficient of 88.27%.
  • Ablation experiments confirmed significant performance gains from the proposed modifications.
  • The model demonstrated superior performance compared to existing segmentation models.

Conclusions:

  • The developed deep learning method effectively segments COVID-19 lung lesions in CT images.
  • This approach can significantly assist clinicians in evaluating and analyzing patient conditions.
  • The study highlights the potential of AI in improving COVID-19 diagnostics and patient care.