Radiologists use artificial intelligence to analyze more than three million studies

September 26

Computer vision algorithms based on artificial intelligence have helped Moscow radiologists to analyze more than three million studies, including more than a million in the last three months.

"Neural networks began helping Moscow radiologists in their work at the beginning of 2020. If initially the new technology was used only for recognizing signs of coronavirus pneumonia, now the diagnostic capabilities have become wider. Now artificial intelligence helps to recognize the signs of COVID-19, lung cancer, breast cancer, osteoporosis, various lung pathologies, cardiovascular pathologies, and in addition, new algorithms are being tested in five other fields. As consequence, with the expansion of diagnostic capabilities, the number of processed images also increases. In total, since the implementing the new technology in Moscow, artificial intelligence has helped doctors to analyze more than three million studies, more than a million of them - over the past three months, which is several times more than the average ones processed last year," the press service of the Moscow Social Development Complex explained.

Thanks to the integration of artificial intelligence services into the unified radiological information service of the Unified Medical Information and Analytical System (ERIS EMIAS), the time for describing the study has also been reduced - by an average of 30 percent. In emergency cases, the time from sending images to the system for receiving the radiologist's report is only a few minutes. This allows you to make diagnose quickly and help the patient.

Artificial intelligence services can be used by radiotherapists of all medical institutions connected to ERIS EMIAS. The unified storage of medical images receives data from more than 1,300 X-ray diagnostics devices.

Computer vision technology based on artificial intelligence works in real time. Immediately after downloading the study to the general system, neural networks form an additional series of images, where the pathologies found are marked. X-rays are first sent to radiologists, which are more likely to contain signs of pathological changes. When describing, for the doctor, both the original study and the image with the results of its processing by artificial intelligence services are available. The use of algorithms improves the quality and speed of radiodiagnosis, helps doctors to track pathologies and identify groups of patients with a high risk of diseases development.

The implementation of such services became possible thanks to the unified digital healthcare platform, which was developed and is being developed by the Moscow Social Development Complex and the Information Technology Department.


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