COVID-ConvNet: A Convolutional Neural Network Classifier for Diagnosing COVID-19 Infection

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• The COVID-19 pandemic poses a global health challenge.
• Chest radiography is a key screening tool for identifying COVID-19.
• Characteristic radiographic anomalies are observed in COVID-19 patients.

Purpose of the Study:

• To introduce COVID-ConvNet, a deep convolutional neural network (DCNN).
• To assess the efficacy of COVID-ConvNet in detecting COVID-19 from chest X-ray (CXR) scans.

Main Methods:

• A deep learning model, COVID-ConvNet, was designed.
• The model was trained and validated on 21,165 CXR images from the COVID-19 Database.
• Performance was evaluated based on prediction accuracy.

Main Results:

• COVID-ConvNet achieved a high prediction accuracy of 97.43%.
• The model demonstrated superior performance compared to existing related works, exceeding them by up to 5.9% in accuracy.

Conclusions:

• COVID-ConvNet is an effective deep learning tool for COVID-19 detection using CXR.
• The model shows significant potential for improving patient screening and diagnosis in pandemic scenarios.