Study on Multi-Viral Infection on Lungs using Data and Predictive Analysis Techniques

Abstract

This research study has used deep learning techniques (DL) to classify spectrograms of acoustic signals. In addition, this study intends to differentiate spectrograms that contain a cough and those that do not contain a cough, and once it is known that a spectrogram contains a cough, it is possible to understand the underlying disease. The main goal is to obtain a system with better performance than those proposed so far in the literature in the field of cough detection and make a first approximation to the classification of diseases based on audio clips with cough.