CIMNE researcher Alberto Tena publishes a study about COVID detection using machine learning

Published: 22/09/2021

Alberto Tena, member of the research team of the Information and Communication Technologies group of CIMNE, has just published an article in the journal Biomedical Signal Processing and Control titled "Automated detection of COVID-19 cough". This article, published in Open access, is a joint work with the researchers Francesc Clarià and Francesc Solsona, from the Department of Computer Science and INSPIRES of the University of Lleida (UdL).

Easy detection of COVID-19 is a challenge. Quick biological tests do not give enough accuracy. Success in the fight against new outbreaks depends not only on the efficiency of the tests used, but also on the cost, time elapsed and the number of tests that can be done massively. The proposal of the study is to provide a solution to this challenge. The main objective is to design a freely available, quick and efficient methodology for the automatic detection of COVID-19 in raw audio files.

 Automatic identification of cough samples in a raw audio file.
Automatic identification of cough samples in a raw audio file

It is based on automated extraction of time-frequency cough features and selection of the more significant ones to be used to diagnose COVID-19 using a supervised machine learning algorithm.

Random Forest has performed better than the other models analysed in this study. An accuracy close to 90% was obtained.

This study demonstrates the feasibility of the automatic diagnose of COVID-19 from coughs, and its applicability to detecting new outbreaks.