According to the Ministry of Health, the platform will help

[tanjimajuha20]

Oliya Artemova, Deputy Director of the Department of Digital Development and Information Technology of the Ministry of Health of the Russian Federation (Ministry of Health), reported at the Innopolis AI Conference for business (AI IN 2023) on August 18 that additional indicators were included in the agreements with the subjects of the Russian Federation on the provision of subsidies from the federal budget for the creation of a single digital circuit: by 2023, the subjects must implement at least one medical product based on AI, and in 2024 - at least three such solutions. According to her, 51 subjects have already purchased them or are in the final stages of the transaction.

Anton Matvienko, assistant director of the Federal State Budgetary Institution "All-Russian Research and Testing Institute of Medical Equipment" of the Federal Service for Surveillance in Healthcare and Social Development (VNIIIMT Roszdravnadzor), reported that there are a total of 23 registered medical products using artificial intelligence technologies. According to him, they can be divided into three types based on their operating principle: medical image analysis, electronic medical record data analysis, and video stream analysis.

Oliya Artemova explained that the introduction of medical products with AI is being carried out within the framework of the federal project "Creation of a single digital circuit in healthcare based on the Unified State Information System in the Sphere of Healthcare (EGISZ)". According to her, the Ministry of Health will monitor the use of AI solutions in the regions on the Artificial Intelligence Platform, which the department created in 2022 for interaction between solution developers and the medical community.

the medical community hong kong whatsapp resource formulate current clinical tasks, organize the collection and labeling of medical data, publish tasks and datasets created for them.

Igor Shulkin, Deputy Director for Prospective Development of the Diagnostics and Telemedicine Center of the Moscow Department of Health, reported that an experiment on introducing artificial intelligence into healthcare was conducted in the capital. "The Diagnostics and Telemedicine Center of the Moscow Department of Health has extensive experience in introducing neural networks into radiation diagnostics. Since 2020, smart algorithms have been working as part of an experiment on introducing computer vision technologies into medicine. Neural networks have already processed 10.5 million radiation studies. Artificial intelligence services are improving, becoming an increasingly popular tool for radiologist. We already have three integrated services that can detect several pathologies at once in one image. Recently, a neural network learned to recognize adrenal gland formations on chest CT. Thus, one service already identifies up to 10 pathologies in one image," said Igor Shulgin.

Robert Vasiliev, director and founder of Real-V LLC (AI & IT lab Z-union), vice president of the Association of Laboratories for the Development of Artificial Intelligence (ALRII), reported that over three years, AI solutions have processed only 0.01% (10.5 million out of 326 million) of all radiation studies conducted by specialists.

Evgeny Popov, head of the healthcare department at Yandex.Cloud, noted that one of the pressing problems in the field of AI is that artificial intelligence must be trained on an array of depersonalized personal data, but this raises the issue of their quantity and quality, as well as the issue of confidentiality: "Depersonalized data, like personal data, must be protected. At the same time, data is an asset, and a culture of working with it must be built."

"The IT sphere lacks the expertise of doctors, while medicine lacks the expertise and resources of IT. Collaborations arise when doctors and IT specialists agree and create services, but these associations are non-systemic and highly specialized," added Evgeny Popov. In his opinion, individual solutions will not eliminate the main problem - the unification of databases for their subsequent labeling and training of artificial intelligence.