Novel Comparative Study for the Detection of COVID-19 Using CT Scan and Chest X-ray Images

Area of Science:

• Artificial Intelligence
• Medical Imaging
• Infectious Diseases

Background:

• The COVID-19 pandemic necessitates rapid and accurate diagnostic methods.
• Traditional RT-PCR testing for COVID-19 is time-consuming and resource-intensive.
• Emerging machine learning and deep learning techniques offer potential for faster, automated analysis of medical images.

Purpose of the Study:

• To evaluate the efficacy of machine learning models for COVID-19 detection using medical imaging.
• To compare the performance of different classifiers, including deep learning architectures, for COVID-19 diagnosis.
• To determine if chest X-rays and CT scans are suitable for AI-driven COVID-19 detection.

Main Methods:

• Development and evaluation of four classification models: SCovNet (CNN), Resnet18 (CNN), Support Vector Machine, and Logistic Regression.
• Training and testing models on a dataset of 17,599 chest X-ray and CT scan images.
• Comparative analysis of model performance based on accuracy in classifying COVID-19 infection.

Main Results:

• The proposed SCovNet architecture demonstrated superior performance.
• SCovNet achieved an accuracy of approximately 99% on chest CT scans.
• SCovNet achieved an accuracy of approximately 98% on chest X-ray images.

Conclusions:

• Deep learning models, particularly SCovNet, show high potential for accurate and efficient COVID-19 detection from medical images.
• Chest CT scans and X-rays can be effectively utilized with AI for rapid COVID-19 diagnosis.
• AI-powered image analysis offers a promising alternative to conventional diagnostic methods, addressing limitations in speed and resource requirements.