AI to help identify COVID 19 victims through cough

1156

An Artificial Intelligence (AI) based algorithm developed by researchers from Massachusetts Institute of Technology (MIT) is said to help detect COVID 19 victims through cough. And results state that the algorithm is efficient enough to differentiate asymptomatic people with those who haven’t caught up the infection.

 

MIT states that the algorithm was developed after analyzing over 200,000 cough samples from the volunteers and proved successful in tracking down 98.5% of people with COVID-19.

 

Cambridge based private research institute states that goal is to analyze more samples and then produce an app that could act like a pre-screening tool of detecting the Chinese virus.

 

“It can prove extremely fruitful to those who are about to visit a classroom, factory or a government office as it helps detect the Wuhan Virus infected people instantaneously”, said Brian Subirana, the director of MIT’s Auto ID Laboratory.

 

Till date the team was working on a similar project that was being helpful to detect pneumonia, asthma and Alzheimer’s disease in time. And now, cough and vocal samples are being collected to analyze those infected with COVID 19 stated Mr. Subirana.

 

MIT, states that it has started to collect samples from April this year allowing people to record their cough and voice samples to track down those infected and the asymptomatic cases.

 

Note- Corona Virus is a newly discovered virus that spreads from person to person through droplets, cough or sneeze. Usually, it takes anything between 2-14 days for the infected person to show symptoms like cold, fever, soar throat, shortness of breath, and breathing difficulties that can lead to severe complications like pneumonia, respiratory disorders, kidney failures and eventually leading to multiple organ failure and death.

Ad
Naveen Goud is a writer at Cybersecurity Insiders covering topics such as Mergers & Acquisitions, Startups, Cyber Attacks, Cloud Security and Mobile Security

No posts to display