An Intelligent Model for Assessment of Cough in Covid-19 Infected Patients based on Sound to Predict their Clinical Criticality using XGB Algorithm

Cough is a one of the major symptoms of COVID 19 infected patients. This research study underlines automatic, objective, and reliable assessment of cough events to predict severity of infection in patients suffering from novel Corona Virus. The system is self-learning and thus intelligent. In this research paper, a brief survey of an audio-based cough monitoring systems, cough detection and then illustrated the cough sound generating principle. Clinical parameters such as cough frequency, intensity of coughing, and acoustic properties of cough sound, were also analyzed in this paper. Steps of Cough sound processing are also considered. Brief studies of cough sound processing algorithms are also made. Finally, XGB algorithm is chosen as the predictor, due to its superb classification and feature selection ability. Finally, end result of clinical criticality of Covid-19 infected patients is predicted. However for the research purpose, the cough sound recording of Covid-19 cough data set is created.. The trained model was used to prediction. The predicted data is found to be accurate