NCI creates a system for the diagnosis of artificial intelligence for 220 million rubles. It will be built into Uniform State Health Information System

2021

The Ministry of Health creates a single database for AI training used in SPPVR

In early September 2021, it became known about the creation in Russia of a single database of medical data, including the results of studies and analyses. It is designed to train artificial intelligence used in the medical decision support system (SPPBR).

The fact that Ministry of Health of the Russian Federation it creates "a single information base of medical documents, images and instrumental research results for the implementation of systems for supporting medical decision-making, including using technologies," artificial intelligence writes "" Kommersant with reference to the department's digital transformation program until 2023. Annually, more than 1 billion will be allocated from the budget for the implementation of the project, in rubles general, the state will invest 3.49 billion rubles in it.

Ministry of Health develops a single database for AI systems that help doctors make diagnoses

In addition, the Ministry of Health intends to stimulate the creation and registration as medical products of 90 domestic innovative SPPVRs using artificial intelligence technologies developed on the basis of data from a single database of the Ministry of Health. In 2021, there should be two such solutions, in 2022 - 30, in 2023 - already 90. The share of medical organizations using SPPVR should grow from 10% in 2021 to 50% in 2023.

The creation of a single database of medical data for doctors will allow building the most effective treatment algorithms, said Mikhail Belyandinov, co-founder and executive director of BestDoctorov.

Ornament Health AG Development Director Georgy Pirkulov, in a conversation with the publication, emphasized that by September 2021, doctors still rely mainly on their experience, although they lack information to make decisions. According to him, doctors will react skeptically if they are told that "it is possible to find out from the picture of the wound what type of damage this is."^[1]

NCI creates a system for the diagnosis of artificial intelligence for 220 million rubles

On April 21, 2021, it became known that the National Center for Informatization (NCI) created a technological system for making a medical diagnosis and prescribing treatment using machine learning and artificial intelligence algorithms .

This was reported by Vedomosti with reference to the documentation published by Rostec's daughter on the public procurement website. For the implementation of the project, NRI buys specialized software for big data analysis. The initial price of the competition is 220 million rubles.

According to a newspaper source familiar with the developers of the platform, it is being created by order of the Ministry of Health of the Russian Federation and will become part of a single state IT healthcare system (EGISZ).

NCI creates a diagnosis system using AI for 220 million rubles

It is needed so that treating doctors can more accurately make diagnoses and prescribe treatment. This will, in particular, improve the quality of diagnostics in those cities where the qualifications of doctors are lower than in Moscow and large regional centers, the source said, adding that in 2020 more than 23 million documents were transferred by doctors in electronic form to the Uniform State Health Information System.

It is assumed that artificial intelligence will compare the history of the patient with the depersonal data of other patients whose electronic cards are entered into the Uniform State Health Information System. AI will analyze medical histories, tests, X-rays, computer tomograms and with a certain degree of probability to make a diagnosis, as well as recommend optimal therapy.

Boris Zingerman, director of the National Medical Knowledge Base Association, in a conversation with Vedomosti, expressed doubts that a diagnostic system based on artificial intelligence would provide noticeable assistance to doctors and patients. There are a lot of symptoms of certain diseases, the condition of the patients themselves can also be very different - AI so far cannot physically take into account all the nuances that are required during diagnosis, since neural networks are not yet sufficiently trained, he said.^[2]

Notes