Related Experiment Video
Updated: Jul 10, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.2K
An Effective Deep Neural Network for Lung Lesions Segmentation From COVID-19 CT Images
Cheng Chen1, Kangneng Zhou1, Muxi Zha1
1School of Computer and Communication EngineeringUniversity of Science and Technology Beijing Beijing 100083 China.
IEEE Transactions on Industrial Informatics
|November 20, 2023
Summary
A new method for segmenting lung lesions in COVID-19 computed tomography (CT) images improves diagnosis. This automated approach enhances accuracy and shows potential for clinical use in managing the COVID-19 epidemic.
Area of Science:
- Medical imaging analysis
- Artificial intelligence in healthcare
- Pulmonary disease diagnostics
Background:
- Accurate segmentation of lung lesions in COVID-19 computed tomography (CT) images is crucial for quantitative diagnosis and treatment planning.
- Existing methods may face challenges in processing CT images effectively during the COVID-19 pandemic.
Purpose of the Study:
- To introduce a novel, automated segmentation method for lung lesions in COVID-19 CT images.
- To enhance the accuracy and efficiency of quantitative analysis for COVID-19 diagnosis and treatment.
Main Methods:
- Region of interest extraction using a patch mechanism strategy for 3-D network applicability and background removal.
- Development of a 3-D network with an attention model to extract spatial features and enhance target areas.
- Implementation of a combined loss function for improved network convergence, gradient optimization, and training direction.
- Application of data augmentation and conditional random fields for data resampling and binary segmentation.
Main Results:
- The proposed method demonstrated superior performance compared to existing approaches in segmentation tasks.
- Comparative experiments validated the effectiveness and accuracy of the novel segmentation technique.
Conclusions:
- The developed automated segmentation method offers a promising tool for clinical applications in COVID-19 diagnosis and treatment.
- The method's high performance suggests its potential to aid radiologists and clinicians in managing COVID-19 patients through quantitative CT analysis.
