New machine learning method for image-based diagnosis of COVID-19
Mohamed Abd Elaziz1,2, Khalid M Hosny3, Ahmad Salah3
1Faculty of Science, Zagazig University, Zagazig, Egypt.
Plos One
|June 27, 2020
Summary
This study introduces a novel machine learning method using Fractional Multichannel Exponent Moments (FrMEMs) to accurately detect COVID-19 from chest X-rays, achieving high diagnostic accuracy.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- The COVID-19 pandemic necessitates rapid and accurate diagnostic tools.
- Machine learning offers potential for automated analysis of medical images like chest X-rays.
- Existing methods may require further optimization for COVID-19 detection.
Purpose of the Study:
- To propose a novel machine learning (ML) method for classifying chest X-ray images for COVID-19 detection.
- To extract relevant features from X-ray images using Fractional Multichannel Exponent Moments (FrMEMs).
- To optimize feature selection using a modified Manta-Ray Foraging Optimization algorithm.
Main Methods:
- Feature extraction using Fractional Multichannel Exponent Moments (FrMEMs).
- Parallel multi-core computational framework for accelerated processing.
- Feature selection via a modified Manta-Ray Foraging Optimization algorithm based on differential evolution.
Main Results:
- The proposed ML method achieved high accuracy in classifying COVID-19 chest X-rays.
- Accuracy rates of 96.09% and 98.09% were obtained on two distinct COVID-19 datasets.
- The FrMEMs feature extraction and optimized selection proved effective.
Conclusions:
- The developed ML approach demonstrates significant potential for accurate COVID-19 diagnosis from chest X-rays.
- The combination of FrMEMs and optimized feature selection offers a robust diagnostic tool.
- This method can aid in the rapid identification of COVID-19 patients.
