Published on: 2026-06-03
Source: Moscow Government – Government of Moscow –
An important disclaimer is at the bottom of this article.
Scientists of the Center for Diagnostics and TelemedicineMoscow Department of Health(DZM) developed and published two training manuals on the application of artificial intelligence (AI) technologies in radiation diagnostics. The first is “Application of medical devices based on artificial intelligence technologies in radiation diagnostics. Computed tomography of the brain, computed tomography and radiography of chest organs, mammography.” The second is “Simulator of the workplace of a radiologist with AI-based services.”
“Today artificial intelligence becomes an integral part of radiation diagnostics and a full-fledged tool in the work of a radiologist. It is important that specialists not only understand the principles of these decisions but also know how to correctly apply them in the diagnostic process. Educational aids are practice-oriented tools that allow seeing the results of neural networks in a safe learning environment and forming the necessary competencies in specialists – from understanding technologiesof computer vision“Before the use of artificial intelligence in clinical practice. This directly affects the quality of diagnostics and the speed of decision-making,” said the director of the DZM Center for Diagnostics and Telemedicine, the chief freelance specialist in radiology and instrumental diagnostics of DZM, Yuri Vasilyev.
Target audience
The manuals are intended for medical university students, residents, practicing radiologists, specialists in medical informatics, and healthcare organizers. They are aimed at developing practical skills in the use of AI services for the analysis, interpretation, and description of the results of radiological studies, as well as demonstrating how the results of artificial intelligence work are presented in the processing of medical images in key areas of radiological diagnostics.
The educational manual “Simulator of the workplace of a radiologist with services based on artificial intelligence” is recommended by the coordination council in the field of education “Healthcare and Medical Sciences” for use in educational organizations implementing residency programs in the specialty “radiology.” The educational manual “Application of medical devices based on artificial intelligence technologies in radiation diagnostics.”Computed tomography of the brain, computed tomography and radiography of the chest organs, mammography are recommended by the coordination council on education in the field of “Healthcare and Medical Sciences” also for future radiologist doctors.
Experience of clinical trials
The manuals are based on the results of the Moscow experiment on the use of computer vision technologies in trauma care, as well as the experience of clinical trials and the implementation of medical devices with artificial intelligence into practice. The publication presents the basic principles and models of AI application in radiological diagnostics, detailed recommendations for the use of algorithms in the analysis of computer tomography of the brain and thoracic organs, radiography and mammography, as well as examples of the application of computer vision technologies in various areas of radiological diagnostics.
“In recent years, significant experience has been accumulated in Moscow in implementing computer vision technologies in radiology — from scientific research to large-scale application in practical healthcare. Today, neural networks are used in real clinical practice and have become part of the everyday work of a radiologist. This expertise formed the basis of the training manuals. The authors tried not just to describe the capabilities of artificial intelligence, but to show real use cases, including limitations and typical errors.””This approach allows specialists to form a holistic understanding of the role of AI in the diagnostic process,” noted Anton Vladzimirskiy, Deputy Director for Scientific Work at the Center for Diagnostics and Telemedicine of the DZM.
Special attention is paid to both correct and incorrect results of algorithm operation, the evaluation of diagnostic accuracy and effectiveness of AI solutions, as well as their integration with the “MosmedAI” platform. The manual provides tasks and control questions for self-checking and consolidating the acquired knowledge.
What in practice
In Moscow, artificial intelligence has already become a full-fledged part of the ray diagnostics system and helps doctors daily analyze medical images. Today, more than 60 computer vision services are used in the metropolitan healthcare system across 45 clinical directions. With their help, more than 30 million studies have already been processed. Neural networks help detect signs of lung cancer, stroke, ischemic heart disease, osteoporosis, pneumonia, aortic aneurysms, and other pathologies on mammograms, computed tomography, and magnetic resonance imaging, as well as on radiological studies.For certain areas, the accuracy of AI services exceeds 95 percent.
The use of artificial intelligence technologies allows for a significant reduction in the workload of radiologists and speeds up diagnostics. Thanks to the implementation of AI, the average time for image description has decreased by approximately 30 percent. At the same time, the final decision always remains with the specialist: neural networks act as a digital assistant that suggests what potentially significant changes to pay attention to and reduces the risk of missing pathologies.An important advantage of the Moscow model is that several developers of AI solutions are connected to the system at once — this ensures constant development and improvement of services.
A separate direction has become the digital platform “MosmedII”, created by order of the President of Russia. It opened access for medical organizations across the country to Moscow AI solutions for the analysis of radiological examinations. Today, the platform is used by more than two thousand medical institutions in 75 Russian regions. Thanks to this, doctors can receive additional expert opinions when interpreting images, increase the speed of diagnostics, and implement modern technologies even in conditions of personnel shortage.
The Diagnostic and Telemedicine Center of the Moscow City Health Department is a leading scientific and practical organization within the structure of the capital’s social development complex, founded in 1996. It specializes in the implementation of artificial intelligence technologies in medicine, the development of radiation diagnostics, the organization of the operation of departments in medical institutions, the conduct of scientific research, and the training of medical personnel.
The development of AI in Moscow medicine is not limited to radiology alone. Artificial intelligence is also applied at the outpatient stage. For example,system for supporting clinical decision makinganalyzes patient complaints and offers therapists and general practitioners three most likely preliminary diagnoses. The specialist can choose one of the options or make their own. The system already determines 95 percent of the most common diseases, with its help more than 15.6 million preliminary diagnoses have been made. Then the service generates a list of necessary examinations and consultations with narrow specialists to confirm the diagnosis. The system also helps doctors in making a final diagnosis for the patient. Artificial intelligence analyzes the information from himelectronic medical recordfor the last two years, as well as complaints and anamneses filled in the current protocol, and generates its own version of the diagnosis. At the same time, the system acts as an assistant, the final decision always remains with the doctor.
In addition, another AI service generates a brief summary for the doctor from the patient’s electronic medical record, including the history of their visits, tests, prescriptions, and emergency calls related to complaints reported by the patient during the online appointment. This allows the specialist to prepare in advance for the appointment and make it more targeted and effective. Since the launch, the AI service has generated over 23.8 million such summaries.
The creation and operation of a digital platform for providing residents with government services in various sectors of the social sphere corresponds to the objectives of the national project“Economy of data and digital transformation of the state”. More about the national project of Russia and the contribution of the capital can be found atspecial page.Â
Get the main news quickly from the official channel of the city of Moscow in the messenger“Max” и“Telegram”.Â
Please note; this information is raw content obtained directly from the information source. It represents an exact report of what the source claims and does not necessarily reflect the position of MIL-OSI or its clients.