Murashko announced a reduction in mortality in clinics by up to 40% due to AI

The use of artificial intelligence (AI) in medical institutions can help reduce patient mortality by up to 40%. This was announced on October 28 by Russian Minister of Health Mikhail Murashko at the Russia—Africa forum.

"Those medical clinics that are actively using artificial intelligence today are already producing very good results in terms of their effectiveness in reducing mortality. <...> Using artificial intelligence, a number of clinics can provide up to 40% effectiveness in reducing mortality rates in the clinic," the minister noted, speaking at the session "Thrombosis: Challenges of our time. Russia—Africa Dialogue".

At the same time, the head of the Ministry of Health stressed that such results should still be treated with caution, despite their encouraging nature. At the same time, he explained that one of the advantages of AI is the ability to analyze large amounts of medical information and identify minor deviations that a doctor may consider accidental. However, the recurrence of such changes or their combination with other factors may indicate the need for additional examination of the patient.

In addition, Murashko pointed out the possibilities of using AI to improve the management of medical institutions, including the organization of patient routing. According to him, a new program is currently being developed that will allow specialists across the country to send patients for treatment to federal medical centers or specialized institutions to receive specialized care.

Neonatal card: who saves premature babies in Russia

Leading experts spoke about new technologies and challenges in the profession.

Moscow Mayor Sergey announced on September 2 that the Moscow-based digital platform MosMedIA is now used by medical organizations in 75 regions of Russia to analyze radiation research using AI. The platform provides 18 AI services to the regions for the analysis of computed tomograms, radiographs, fluorograms and mammograms. They are able to detect up to 14 different pathologies in one study, and image processing in most cases takes no more than five minutes.