Related Experiment Video
Updated: Oct 9, 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
14.4K
D2-CovidNet: A Deep Learning Model for COVID-19 Detection in Chest X-Ray Images
Xin Wang1, Yiyang Hu1, Yanhong Luo2
1School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha 410114, China.
Computational Intelligence and Neuroscience
|December 20, 2021
Summary
A new deep learning model, D2-CovidNet, aids in diagnosing Coronavirus disease 2019 (COVID-19) using chest X-rays. This lightweight network achieves high accuracy, enabling faster and more precise detection of COVID-19 pneumonia.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Vision
Background:
- The global spread of Coronavirus disease 2019 (COVID-19) necessitates rapid and accurate diagnostic tools.
- Chest X-ray imaging is a key modality for identifying COVID-19 pneumonia, but detection accuracy requires improvement.
Purpose of the Study:
- To develop a novel, lightweight convolutional neural network (CNN) for enhanced COVID-19 detection from chest X-ray images.
- To improve the accuracy and efficiency of COVID-19 diagnosis through advanced feature extraction techniques.
Main Methods:
- Introduction of a dual-path multiscale feature fusion module and a dense depthwise separable convolution module.
- Design and implementation of D2-CovidNet, a CNN integrating these novel modules for image analysis.
- Validation on two public datasets to assess diagnostic performance.
Main Results:
- D2-CovidNet achieved an overall classification accuracy of 94.56%, with high precision (95.14%), sensitivity (94.02%), specificity (96.61%), and F1-score (95.30%).
- The model demonstrated exceptional performance for COVID-19 detection, with precision at 98.97%, sensitivity at 94.12%, and specificity at 99.84%.
- D2-CovidNet exhibits a reduced computational load and fewer parameters compared to existing methods.
Conclusions:
- D2-CovidNet offers a computationally efficient and accurate approach for diagnosing COVID-19 from chest X-rays.
- The proposed network assists medical experts in making quicker and more reliable diagnoses.
- The integration of specialized modules enhances feature sensitivity for improved pneumonia detection.
