New unique city service assesses pneumonia severity using blood tests

December 15, 2020

The city has teamed up with researchers from  Lomonosov Moscow State University to create a neural network called the CAT Scan Calculator for evaluating the extent of patients’ lung problems.

The trained neural network will help doctors predict the probability of light (CAT 0-1), medium (CAT 2) or severe (CAT 3-4) pneumonia cases and decide on subsequent treatment options. In some cases, if the calculator predicts a light case of pneumonia, the patients will not need to have a CAT scan. All other patients will be immediately hospitalised and undergo CAT or X-ray scans, as well as intensive treatment.

The neural network was trained by comparing blood test results, saturation levels and overall clinical symptoms of COVID-19 plus pneumonia cases with CAT scans. Moscow doctors have treated over 500,000 people since March 2020. A large amount of data has been collected. The service will continue to improve and expand its expertise as new data appears.

The CAT Scan Calculator has already been integrated with the Uniform Medical Information and Analytical System (UMIAS), which is accessible to doctors from Moscow and other regions as well.

Digital healthcare services have been developing rapidly during the pandemic. Moscow’s doctors have been using special software that analyses lung CAT scans for diagnosing pneumonia accompanied by COVID-19 for the past ten months. Artificial intelligence helps doctors to accurately diagnose large numbers of patients.

Other innovations include online case histories. From now on, Muscovites can see the results of their medical tests, including coronavirus and COVID-19 antibody tests, medical appointments and ambulance calls. Over 1.7 million people have used the service to date.

For more information about COVID-19 prevention and treatment, go to this special project.


Latest Events
Event finished

Sports and Recreation at VDNH: Fun, Diverse, and Safe

June 1December 31, 2020

Aida Garifullina

February 29, 2020
If you continue to use our website, you are agreeing to accept the use of cookies on your device. Cookie files ensure the website’s efficiency and help us provide you with the most interesting and relevant information. Read more about cookie files.
Accept ccokies