A screening tool for covid-19 based on voice analysis

A screening tool for covid-19 based on voice analysis


A team of Iranian researchers from Alzahra University (Dr. Mandana Nourbakhsh, Dr. Maral Asiaee), Baqiyatallah University of Medical Sciences (Dr. Amir Vahedian Azimi), Science Information Database (SID)(Abdolsamad Keramatfar) and Jundishapour University of Medical Sciences (Seyyed Shahab Atashi) have developed a software  -as a screening tool- that distinguishes individuals with COVID-19 from healthy ones based on the acoustic analysis of their voices. Dr. Mandana Nourbakhsh, the project's leader, explains that, since COVID-19 is mostly considered as a respiratory disease and many of its symptoms are associated with the larynx and/or the lungs infections, those acoustic parameters that represent these organs' functions are very likely to be affected by the Corona virus; Therefore, these researchers first evaluated and compared the acoustic parameters of voice between healthy participants and those infected with COVID-19. They found out that there were significant differences in values of multiple parameters related to voice quality between the COVID-19 patients and the healthy participants. The results are published here. Based on these findings, the team, then used artificial intelligence (AI) to design a software that differentiates covid-19 patients form healthy individuals, solely by analyzing their voice. Their model correctly identifies coronavirus patients 90.5% of the time.  This screening tool, is now available online. It helps people to make sure that their respiratory system is healthy without going to medical centers, resorting to invasive methods and spending lots of money and time. The software advises people to go to medical centers for further examination, if some signs of the disease are found in their voices. The software is yet to be approved by the ministry of health and medical education of Iran.

summary-address :
Your Rating
Average (1 Vote)
The average rating is 5.0 stars out of 5.


Last news

Mon, 14 Dec 2020 05:15