Respiratory Pattern Analysis for COVID-19 digital Pre-screening using Machine Learning and Deep Learning

Corona Virus (COVID-19) is a highly infectious respiratory disease, and it has been announced as a global pandemic by the World Health Organization (WHO). This virus has spread worldwide, affecting various countries until now, causing 4 million deaths worldwide. To tackle this public health crisis, medical professionals, researchers are working relentlessly, applying different techniques and methods. AI-based classification algorithms are used as one of the tools in diagnosing covid-19. Some common COVID symptoms seen in patients include cough, headache, cold, breathing problem, high temperature, chest congestion, and loss of sense of smell [2]. In terms of diagnosis by using a digital stethoscope, respiratory sound has been recognized as an indicator of health. Our research focuses on human respiratory sound parameters such as cough, voice, and breath. We will provide an audio-based Machine Learning (ML) and Deep Learning (DL) model for COVID-19 digital screening based on audio data of COVID patients. This work will open the door to investigate further how automatically analyzed respiratory patterns could be used as pre-screening signals to aid COVID-19 diagnosis and control the outbreak.