Covid-19: New AI tool can predict if infected person will die or survive

Prashasti Awasthi Mumbai | Updated on February 07, 2021

File photo   -  REUTERS

Researchers at the University of Copenhagen developed a new artificial intelligence tool that can 90 per cent predict whether a person infected with COVID-19 will die or survive.

The study, published in the journal Scientific Reports -- Nature, suggested that body mass index (BMI), gender, and high blood pressure are among the most heavily weighted factors. This can be used to develop a model that can predict the likeliness of dying from the virus.

The tool will also help in determining the priority groups for the vaccine inoculation.


The findings of the study were based on patient data from the Capital Region of Denmark and Region Zealand. The findings demonstrated that artificial intelligence can, with up to 90 per cent certainty, determine whether an uninfected person who is not yet infected will die of Covid-19 or not if contracted the virus.

Once admitted to the hospital with Covid-19, the computer can also predict with 80 per cent accuracy whether the person will need a respirator.

Professor Mads Nielsen of the University of Copenhagen's Department of Computer Science said: "We began working on the models to assist hospitals, as, during the first wave, they feared that they did not have enough respirators for intensive care patients. Our new findings could also be used to carefully identify who needs a vaccine."

For the study, the researchers fed a computer program with health data from 3,944 Danish Covid-19 patients.

"Our results demonstrate, unsurprisingly, that age and BMI are the most decisive parameters for how severely a person will be affected by Covid-19. But the likelihood of dying or ending up on a respirator is also heightened if you are male, have high blood pressure or neurological disease," explained Mads Nielsen.

Published on February 07, 2021

Follow us on Telegram, Facebook, Twitter, Instagram, YouTube and Linkedin. You can also download our Android App or IOS App.

This article is closed for comments.
Please Email the Editor

You May Also Like