Related Experiment Video
Updated: Oct 20, 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
Detection of novel coronavirus from chest X-rays using deep convolutional neural networks
Shashwat Sanket1, M Vergin Raja Sarobin1, L Jani Anbarasi1
1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India.
Multimedia Tools and Applications
|September 13, 2021
Summary
A new deep learning model, CovCNN, uses chest X-rays for rapid COVID-19 detection. This convolutional neural network (CNN) approach achieves 98.4% accuracy, aiding medical professionals in diagnosing coronavirus infections quickly.
Area of Science:
- Artificial Intelligence
- Medical Imaging
- Computational Biology
Background:
- The COVID-19 pandemic necessitates rapid and accurate diagnostic tools due to increasing global infections.
- Traditional laboratory testing for COVID-19 is often slow, costly, and requires specialized facilities.
- The need for efficient diagnostic methods is critical for managing patient care amidst high workloads.
Purpose of the Study:
- To propose a novel Convolutional Neural Network (CNN) model, named CovCNN, for the expedited diagnosis of COVID-19.
- To leverage chest X-ray images for automated COVID-19 detection, assisting medical practitioners.
- To develop a fast and reliable diagnostic system to address the limitations of current testing methods.
Main Methods:
- Development of a deep CNN architecture (CovCNN) with multiple CNN layers.
- Utilization of depthwise convolution with varying dilation rates for feature extraction from X-ray images.
- Training and evaluation of the model using a dataset of 657 chest X-rays (219 COVID-19 positive, 438 non-COVID-19).
Main Results:
- The proposed CovCNN model achieved a highest classification accuracy of 98.4% in detecting COVID-19 from chest X-rays.
- Performance evaluation included analysis of loss and accuracy curves using various pre-trained models.
- The model demonstrated significant potential in accurately identifying COVID-19 positive cases.
Conclusions:
- The CovCNN model offers a promising, highly accurate, and rapid solution for COVID-19 diagnosis using chest X-rays.
- This AI-driven approach can significantly assist healthcare professionals in high-volume diagnostic scenarios.
- The study highlights the effectiveness of deep learning in medical image analysis for infectious disease detection.
