Detection and Diagnostic Approach of COVID-19 Based on Cough Sound Analysis
- 1 University of Human Development, Iraq
- 2 Al-Qalam University College, Iraq
- 3 Tikrit University, Iraq
- 4 University of Anbar, Iraq
Abstract
Coronavirus (COVID-19) started at the end of 2019 and then spread out around the world as a pandemic at the beginning of 2020. At that time, researchers began to work on detecting and diagnosing this virus, where many methods have been applied for this reason. This study focuses on how to diagnose coronavirus through patients’ cough. Accordingly, real samples were taken from people infected by the coronavirus and others, who are suffering from some respiratory diseases. The cough of a person with coronavirus is characterized by its dryness and differs from other cough sounds through a set of factors that are considered for study and analysis through this study. Among these factors is the sound energy, which is found to be the most effective factor and hence implemented as a key indicator for COVID-19 detection. The discrete wavelet transform is the adopted method to realize the detection process via approximation and the analysis of coefficients details. The obtained results show acceptable detection accuracy for the considered samples. Minor mismatching in the detection process is noticed during the procedure, which is mainly due to some patients being infected with the respiratory diseases that exhibit similar symptoms.
DOI: https://doi.org/10.3844/jcssp.2021.580.597
Copyright: © 2021 Muzhir Shaban Al-Ani, Thabit Sultan Mohammed, Awni Ismail Sultan, Hasan Ismail Sultan, Khattab M. Ali Alheeti and Karim Mohammed Aljebory. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
- 3,799 Views
- 1,406 Downloads
- 0 Citations
Download
Keywords
- Corona Virus
- COVID-19
- Cough Sound
- Signal Processing
- Statistical Analysis
- Feature Extraction