An AI diagnostic app to predict the probability of getting COVID-19

Recent research published in Nature Medicine summarizes work by the Kings College London, Massachusetts General hospital and ZOE (health science company) to predict the likelihood of infecting novel coronavirus based on their existing symptoms. The AI uses the database from the “COVID symptom study” app, which has already been downloaded by over 3.3 million people to record their health status. Symptoms like cough, fever, etc. and other health conditions are updated by the users in the app for the

The AI tool predicts with 80% accuracy the probability of getting infected by COVID-19. The study takes in demographic and four vital symptoms: fatigue, the habit of skipping meals, severe and persistent cough and the most critical loss of smell or taste (anosmia) based on the database of ~2.5 million downloads from people in the UK and the US. A stunning insight is that anosmia is a stronger predictor of COVID-19 than fever, a fact not known globally. The entire research and findings have been published in “Nature Medicine”.

 

Intern, Vidula

Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Join the kick-ass newsletter for a delightful freebie and the latest content from The Middle Road; including exclusive research, posts, tweets, videos, and much more.

Join the kick-ass newsletter for a delightful freebie and the latest content from The Middle Road; including exclusive research, posts, tweets, videos, and much more.

This website uses cookies to ensure you get the best experience on our website.