Respiratory Pattern Analysis for COVID-19 Digital Screening Using AI Techniques
Abstract
Corona Virus (COVID-19) is a highly contagious respiratory disease that the World Health Organization (WHO) has declared a worldwide epidemic. This virus has spread worldwide, affecting various countries until now, causing millions of deaths globally. To tackle this public health crisis, medical professionals and researchers are working relentlessly, applying different techniques and methods. In terms of diagnosis, respiratory sound has been recognized as an indicator of one’s health condition. Our work is based on cough sound analysis. This study has included an in-depth analysis of the diagnosis of COVID-19 based on human cough sound. Based on cough audio samples from crowdsourced COVID data, we develop an audio-based framework, deploying traditional Machine Learning (ML), Resampling multiclass ML approach, Cost-Sensitive Multiclass ML, and Multiclass Deep Learning (DL) approaches for COVID-19 digital screening. Our experimental results indicate that the resampling Multiclass ML approach shows the best result for COVID-19 digital prescreening with an AUC of 78.77%. To the best of our knowledge, this is the first COVID-19 detection tool that uses such diverse crowdsourced and largest physician annotated COVID data that uses patients’ cough sound data to predict the presence of COVID-19 in those patients by applying multiple multiclass data balance techniques for AI algorithms. Our proposed novel framework and the developed tool will assist in a) automating COVID-19 digital pre-screening, b) making the COVID test more accessible and cost-effective, c) helping to detect an infected individual before a physical COVID test, and d) reducing the risk of infecting others.