Business Today
Loading...

Coronavirus treatment: This AI-based app will detect if you are asymptomatic

A user could log in daily and cough into their phone. The user will instantly receive information on whether they are infected from COVID-19 or not and then confirm it with a formal test

twitter-logoBusinessToday.In | October 30, 2020 | Updated 10:06 IST
Coronavirus treatment: This AI-based app will detect if you are asymptomatic
According to scientists people who are asymptomatic, cough in a different way than healthy individuals

Scientists have found a way to distinguish between an asymptomatic coronavirus person and a healthy individual.  A team of researchers from Massachusetts Institute of Technology (MIT)  in the US has developed an Artificial intelligence (AI) model that will record coughing and show results on smartphones.

According to scientists people who are asymptomatic, cough in a different way than healthy individuals.

HOW THE AI MODEL WILL WORK?

A user could log in daily and cough into their phone. The user will instantly receive information on whether they are infected from COVID-19 or not and then confirm it with a formal test.  Scientists are working on incorporating the model into a user-friendly app, which if approved and adopted, could identify people who are likely to be asymptomatic for COVID-19.

THE STUDY

In a paper published in the IEEE Journal of Engineering in Medicine and Biology, researchers stated that they tested thousands of samples of cough, including samples of asymptomatic patients, COVID-19 patients as well as healthy individuals.

They created a website where people were able to record a series of coughs through a cellphone or other web-enabled device. During the study, volunteers were also told to fill out a survey of symptoms they were experiencing, and whether they were diagnosed through an official test, by a doctor's assessment of their symptoms, or if they were self-diagnosed.

Researchers collected more than 70,000 recordings, each containing several coughs, amounting to some 200,000 forced-cough audio samples. The team used 2,500 COVID-associated recordings, along with 2,500 more recordings that they randomly selected from the collection to balance the dataset.

They used 4,000 of these samples to train the AI model. The remaining 1,000 recordings were then fed into the model to see if it could accurately discern coughs from COVID patients versus healthy individuals.

The researchers were able to pick up patterns in the four biomarkers -- vocal cord strength, sentiment, lung and respiratory performance, and muscular degradation -- that are specific to COVID-19.

Prior to the coronavirus outbreak, research groups already had been training algorithms on cellphone recordings of coughs to accurately diagnose conditions such as pneumonia and asthma.

Also rtead: 'Pick-eye' or 'pee-chay': US senators fail to pronounce Google CEO Sundar Pichai's name

Also read: BMC spends Rs 82 lakh in Kangana Ranaut's office demolition case

Youtube
  • Print

  • COMMENT
BT-Story-Page-B.gif
A    A   A
close