INASNET: Automatic identification of coronavirus disease (COVID-19) based on chest X-ray using deep neural network

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Disease Diagnostics

Background:

• COVID-19, caused by SARS-CoV-2, presents symptoms similar to pneumonia and influenza, complicating diagnosis.
• Accurate and rapid testing is crucial for community transmission control and patient isolation.
• Deep learning has shown effectiveness in medical image analysis for disease identification.

Purpose of the Study:

• To develop and evaluate a deep learning model for identifying COVID-19 infection from X-ray images.
• To create a cost-effective and efficient diagnostic tool for COVID-19.
• To differentiate COVID-19 from pneumonia and normal cases using chest X-rays.

Main Methods:

• A deep learning model, Inception Nasnet (INASNET), was proposed.
• INASNET integrates InceptionNet and Neural Network Architecture Search (NAS).
• The model was trained and tested on X-ray images to classify normal, pneumonia, and COVID-19 cases.

Main Results:

• The INASNET model demonstrated the ability to classify X-ray images into normal, COVID-19, or pneumonia categories.
• The proposed method offers a cheaper alternative to existing COVID-19 testing kits.
• Continuous analysis and evaluation aim to improve accuracy and reduce false negatives.

Conclusions:

• The INASNET model provides an economically viable and efficient method for COVID-19 diagnosis using X-ray imaging.
• This AI-driven approach can significantly aid frontline workers in managing the pandemic.
• Faster and more accurate diagnoses through this method can help control the spread of COVID-19.