Non-iterative learning machine for identifying CoViD19 using chest X-ray images.

Area of Science:

• Medical Imaging
• Machine Learning
• Computational Biology

Background:

• COVID-19, caused by SARS-CoV-2, led to a global pandemic with significant health impacts.
• The Omicron variant demonstrated high transmissibility, contributing to widespread infections.
• Pneumonia is a dangerous symptom of COVID-19, necessitating rapid and accurate detection methods.

Purpose of the Study:

• To automate COVID-19 detection using machine learning techniques on chest X-ray images.
• To evaluate the efficacy of higher order statistics in analyzing chest X-ray disturbances for disease identification.
• To compare iterative and non-iterative models for speed and accuracy in COVID-19 diagnosis.

Main Methods:

• Experimentation on COVID-19 chest X-ray images.
• Application of higher order statistics for image analysis.
• Development and comparison of iterative and non-iterative machine learning models.

Main Results:

• A non-iterative model achieved a high accuracy of 96.64%.
• The non-iterative model demonstrated superior speed for patient testing compared to iterative models.
• The proposed method showed efficacy when compared to existing state-of-the-art and iterative techniques.

Conclusions:

• Machine learning, particularly with higher order statistics on chest X-rays, offers a promising avenue for automated COVID-19 detection.
• Non-iterative models provide a faster alternative for rapid patient screening.
• The developed method is effective and efficient for COVID-19 diagnosis.