Content |
Main article: Artificial intelligence (AI), Artificial intelligence (AI)
Artificial Intelligence in Medicine
Main article: Artificial Intelligence in Medicine
2023
GOST developed by metropolitan doctors for artificial intelligence services in clinical medicine came into force
Developed with the participation of Moscow radiologists GOST, necessary for the safe design and maintenance of systems artificial intelligence in clinical, came into force. to medicine It establishes the basis of life cycle processes. neuronets This is the 11th domestic standard. It will be used by developers of artificial intelligence services, laboratory employees during technical tests and specialists of medical organizations in clinical trials. This was announced on September 20, 2023, Anastasia Rakova by the acting Deputy Mayor for Moscow Social Development.
 | The rapid development of artificial intelligence technologies, which is underway in Moscow healthcare, raised the question of the need to adopt norms governing its use. Special attention is paid to this issue: metropolitan specialists have been working on standards for the third year. The document defines the requirements for each life cycle process of artificial intelligence systems. The use of standards will ensure the introduction of only proven, reliable solutions in practical medicine, will help in the development of new neural networks and testing existing ones, - said Anastasia Rakova. |  |
This standard complements the existing GOST IEC 62304-2022. The authors added to the set of life cycle processes of software for medical devices the need to monitor the artificial intelligence system at all stages related to operation.
To develop unified concepts and norms, the capital united several dozen specialists: software developers, doctors, professors, researchers. Among the authors of the standard are specialists from the Center for Diagnostics and Telemedicine of the Moscow Department of Health.
| | On the basis of the Center for Diagnostics and Telemedicine of DZM, a scientific Experiment is being carried out to introduce computer vision into radiation diagnostics. The accumulated experience and scientific results of the experiment made it possible for our specialists to begin work on the creation of national standards. 11 GOSTs have been developed, 10 of which entered into force and operate on January 1, 2023, at the moment 3 more draft standards have been publicly discussed, "said Yuri Vasiliev, chief radiologist of Moscow, director of the Center for Diagnostics and Telemedicine of the DZM . | |
The initiative to create national standards for the introduction of artificial intelligence systems in clinical medicine belongs to the specialized subcommittee of the technical committee for standardization "Artificial Intelligence" of Rosstandart, organized on the basis of the Center for Diagnostics and Telemedicine of the Moscow Department of Health.
Two more GOSTs on AI testing in clinical medicine came into force in Russia
On January 1, 2023, standards (GOST R 59921.8-2022 and GOST R 59921.9-2022) entered into force in Russia, establishing general requirements for artificial intelligence systems in medicine and quality management systems. This was reported to Zdrav.Expert on January 11, 2023 by representatives of the Center for Diagnostics and Telemedicine of DZM (NPCC DiT DZM). According to them, the adopted unified standards are designed to help in the development of new neural networks and testing existing ones, and will also allow only the best practices to be introduced into practical medicine.
Both GOSTs were developed by the Center for Diagnostics and Telemedicine of the Moscow Department of Health. The initiative to create national standards for the introduction of artificial intelligence systems in clinical medicine belongs to the specialized subcommittee 01 of the Technical Committee for Standardization 164 "Artificial Intelligence" of Rosstandart, organized on the basis of the NPCC DiT DZM. To develop uniform concepts and norms for all participants in the process, the capital united several dozen specialists - software developers, doctors, professors, researchers.
The documents establish the general provisions of the set of national standards "Artificial Intelligence Systems in Clinical Medicine." In addition, they determine the requirements for methods of testing artificial intelligence systems (SI) for the analysis of medical images.
So, GOST R 59921.8-2022 ("Artificial intelligence systems in clinical medicine. Part 8. Guidelines for the application of GOST ISO 13485-2017 ") will offer organizations a guide to the application of GOST ISO 13485 to processes related to the order, supply, development, operation of SI.
In turn, GOST R 59921.9-2022 ("Artificial intelligence systems in clinical medicine. Algorithms for data analysis in clinical physiology. Test methods. General requirements ") applies to Data Mining those obtained by research methods in clinical physiology, it will determine the organization of work on the standardization of tests algorithms and SIS used for processing and analysis. data
First of all, GOSTs of the Russian Federation will be used by developers of artificial intelligence services, laboratories in technical tests and medical organizations in clinical trials.
| | "We are seeing a tremendous breakthrough in artificial intelligence in Moscow medicine. Last year, the first GOSTs of the Russian Federation appeared for testing neural networks in clinical medicine, now new ones have come into force. Thus, 10 GOSTs of the Russian Federation, which were developed by the capital's radiologists, have already been adopted and applied, "said Anastasia Rakova, Deputy Mayor of Moscow for Social Development. | |
According to her, experts create standards for artificial intelligence systems in healthcare, which are not yet in the world, relying on their unique experience and scientific developments. Such standards increase confidence in modern technologies and make all processes more transparent and understandable at the development stage, during registration, and during operation. In turn, this helps to speed up the introduction of smart solutions, as well as get better and more predictable tools for caring for the health of patients, emphasized Anastasia Rakova.
As of January 2023 metropolitan medicine , more than 40 artificial intelligence services in 17 different areas of research have been introduced. Doctors use neural networks to process fluorography,, mammographies radiography, and. computed tomography magnetic resonance imaging Artificial intelligence helps to find signs of,, lung cancer COVID-19 osteoporosis spine, aorta, aneurysms coronary disease,,, hearts stroke pulmonary hypertension hydrothorax, as well as other breast cancer flat-footedness diseases on radiation studies.
2022
Moscow radiologists have developed GOST for testing AI in clinical medicine
Moscow doctors - employees of the Center for Diagnostics and Telemedicine of the Department of Health - have developed a standard regulating the conduct of clinical trials of artificial intelligence systems. Previously, such a document did not exist for these systems. On August 2, 2022, Zdrav.Expert was informed about this by representatives of the Center for Diagnostics and Telemedicine (NPCC DiT DZM). Read more here.
Russia is the first in the world to adopt AI standards for healthcare
In February 2022, it became known that Russia was the first in the world to adopt standards in the field of artificial intelligence in healthcare. According to the director of the Institute of Digital Medicine, head. Department of Information and Internet Technologies of the First Moscow State Medical University named after I.M. Sechenov Georgy Lebedev, the first five standards will come into force during 2022, including:
- "Intelligent methods of medical data processing. Basic Provisions ";
- "AI systems in clinical medicine - program, methodology of clinical trials";
- "Standard for Change Management in AI Systems with Continuous Learning."
It is noted that by February 9, 2022, another 58 standards for medical AI systems with continuous training and about 60 are being prepared in Russia, which determine the requirements for the interaction of medical devices and devices with electronic medical records, as well as interaction with the digital health care system. Two technical committees for standardization are engaged in the preparation of documents, one of which is headed by Georgy Lebedev, the other is the director of the Scientific and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Sergei Morozov Department of Health.
At the same time, by February 2022, there are no international standards in the field of artificial intelligence for healthcare, only recommendations for their development have been adopted: an international group of specialists with the participation of Russian experts has been created to prepare documentation. According to the decision of the World Health Organization, the authority to develop is assigned to one of the technical committees of this organization.
By April 1, 2022, the Government of the Russian Federation, on behalf of President Vladimir Putin, will have to prepare a report on improving the mechanisms for applying experimental legal regimes in the field of digital innovation. In particular, the Cabinet was instructed to pay special attention to the need to identify those responsible for causing harm as a result of the use of AI-based solutions.[1]
2021: AI in medicine standardized
On October 20, 2021, it became known that the technical committee for standardization published the final versions of three draft national standards for the use of AI in medicine. Market players assess the standards positively and believe that they can be worked with.
The standards include several guests. The first standard is GOST R "Artificial Intelligence Systems. Artificial intelligence systems in clinical medicine. Part 1. Clinical trials. " It establishes the general requirements for conducting clinical trials and evaluating artificial intelligence systems (A.I.) to determine the level of safety and clinical significance of the A.I. output, the general principles of applying A.I. accuracy assessment indicators.
The second standard is GOST R "Artificial Intelligence Systems. Artificial intelligence systems in clinical medicine. Part 4. Evaluation and monitoring of operational parameters. " The standard establishes general requirements for the assessment and control of the operational parameters of the artificial intelligence system during commissioning and periodic control, which will allow for an unambiguously interpreted assessment of the characteristics and parameters of the AI.
The third standard is GOST R "Artificial Intelligence Systems. Artificial intelligence systems in clinical medicine. Part 5. Requirements for the structure and order of application of the data set for training and testing algorithms. " The standard establishes general requirements for the structure and order of application of datasets that are used to train and test artificial intelligence systems.
Sergey Voinov, director of acceleration in digital medicine at the Skolkovo Foundation's biomedical technology cluster, believes that Technical Committee for Standardization No. 164 was formed from a good expert group on the development, refinement and adaptation of those standards that existed in the world for the use of artificial intelligence in medicine.
| | The final version of these standards, which were transformed into GOST, reflects the objective reality of the cases that we see in the artificial intelligence market. This is a good step towards harmonizing the standards for the use of artificial intelligence. And here we take a good expert position: part of our group is working on the development of standards, ISO that is, international. Work was also carried out to adapt these international standards to the realities, Russian health care- said Sergey Voinov. | |
| | Of course, they will need updating, because artificial intelligence technologies are a new field, and there are constantly products, technological solutions, and, accordingly, this affects an increasingly wide range of medical processes. At some point in time, a sufficient number of technology cases will accumulate when these standards need to be updated. But for October 2021, I think it's good that they came out, and it will be convenient for many market players to focus on them, "said Sergei Voinov. | |
In accordance with the promising work plan of TK 164, by 2027 about 50 standards in the field of AI in healthcare should be developed, notes Alexander Rozhkov.
| | Primary standards for an area such as AI should be term review and technology classification. In other words, an ontological model of AI should be drawn up in relation to the field of health care. In fact, it is necessary to carry out a lot of research work to study existing practices and promising areas of using AI in healthcare, "said Alexander Rozhkov. | |
{{quote "Over the past year and a half, many HealthTech and COVID-Tech startups have appeared on the market, in the solutions of which AI technologies are somehow used. A large number of potential directions, as well as the level of maturity of AI technologies, can be identified in the framework of a detailed consideration of the "road map" "Helsnet" of the National Technology Initiative and the "road map" of the development of the "end-to-end" digital technology "of Neurotechnology and artificial intelligence, - said Alexander Rozhkov. }}
Of course, when developing standards, it is necessary to take into account many organizational, technological, ethical and legal aspects, emphasizes Alexander Rozhkov.
| | Including it is necessary to take into account the standards outlined in the Concept for regulating AI and robotics technologies until 2024 - the availability of data sets and image libraries for training neural networks, self-driving and self-learning diagnostic models, priority "second" opinions, responsibility for errors in diagnosing, risk management, methods for assessing the effectiveness of diagnostics, embedding such systems in a single digital health care circuit, ensuring cybersecurity, import substitution, certification and patenting of algorithms/systems/services/platforms based on AI, training personnel to work with systems, etc., - said Alexander Rozhkov, director of product technologies of the group "T1"[2]. | |
2020
Creating Standards
On August 24, 2020, it became known about the development of the first national standard in Russia for artificial intelligence in healthcare. The GOST R project, which will regulate the conduct of clinical trials of medical AI systems (SI), was created by experts from the Center for Diagnostics and Telemedicine of the Moscow Department of Health, together with the Artificial Intelligence in Healthcare subcommittee (part of the Technical Committee No. 164 of Rosstandart).
| | This standard establishes the general requirements for the CT and clinical assessment of the SI to determine the level of safety and clinical significance of the output data of the SI, the general principles of the use of indicators for assessing the accuracy of the SI, is indicated in the document, an excerpt from which is provided by Vademec. | |
As Sergey Morozov, director of the Center for Diagnostics and Telemedicine and chairman of the subcommittee of the Artificial Intelligence in Healthcare PK01/TK164, explained, AI technologies are aimed at optimizing the work of diagnostics, automating routine work, information support and introducing other options that help doctors. The practical use of such systems in medicine is impossible without uniform regulatory and technical norms, he added.
The Ministry of Health notes that, according to the national strategy for the development of artificial intelligence for the period up to 2030, creating conditions that will ensure the use of artificial intelligence technologies in the field of health care is one of the key tasks to improve the standard of living of the population. Issues of trust in medical intelligent systems and barriers associated with the use of AI systems in medical diagnostics can be removed by standardizing the requirements for testing methods for intelligent systems, as well as criteria for compliance of these systems with safety requirements.[3]
Plans to develop the first editions of national standards in the field of AI
On February 26, 2020, RVC told Zdrav.Expert that in 2020 the first editions of national standards in the field of artificial intelligence in healthcare will be developed in Russia. The development will be coordinated by the Technical Committee for Standardization "Artificial Intelligence" (TK 164), created on the basis of RVC.
In accordance with the long-term work plan of the Technical Committee until 2027, it is planned to develop about 50 standards in the field of artificial intelligence in healthcare in certain areas, including general requirements and classification of AI systems in clinical medicine, radiation and functional diagnostics, remote monitoring systems, histology, medical decision support systems, image reconstruction in diagnosis and treatment, big data in healthcare, medical analytics and forecasting systems, educational programs in healthcare.
The editions of the first six national standards will be prepared in 2020, among them:
- Artificial intelligence systems in clinical medicine. Part 1. Clinical trials;
- Artificial intelligence systems in clinical medicine. Part 2. Technical test program and methodology;
- Artificial intelligence systems in clinical medicine. Part 3. Application of quality management to post-training programs. Amendment Protocol; algorithm
- Artificial intelligence systems in clinical medicine. Part 4. Evaluation and monitoring of operational parameters of further-trained algorithms;
- Artificial intelligence systems in clinical medicine. Part 5. Requirements for the structure and procedure of applying a dataset for training and testing algorithms;
- Artificial intelligence systems in clinical medicine. Part 6. General operating requirements.
| | Healthcare is one of those areas where the standardization of artificial intelligence technologies is most relevant. This is explained, firstly, by the danger of the consequences of the use in clinical practice of AI systems with non-guaranteed quality, and secondly, by the high technological maturity of developments in this area. There are several dozen companies in our country alone offering highly advanced intelligent medical information processing systems, and the lack of an adequate regulatory framework is a deterrent to the effective implementation of these systems. said Chairman of the Technical Committee "Artificial Intelligence" Sergey Garbuk
| |
In order to coordinate the unification and standardization of requirements for AI systems in healthcare, as well as the establishment of certification requirements for medical devices, by order of Rosstandart, a subcommittee "Artificial Intelligence in Healthcare" (PK01/TK164) was created as part of TK 164. The subcommittee was formed on the basis of the Center for Diagnostics and Telemedicine. It was headed by the chief freelance specialist in instrumental and radiation diagnostics of the Department of Health of Moscow and the Ministry of Health of Russia in the Central Federal District, director of the Center for Diagnostics and Telemedicine, Doctor of Medical Sciences, Professor Sergei Morozov.
| | National standards will regulate key aspects of the application of artificial intelligence in health care and its role in medical decision-making. Without uniform regulatory and technical norms approved by the professional community on a consensus basis, practical implementation of artificial intelligence in the healthcare sector is impossible. I emphasize that the main emphasis in our work will be on practical health care. We will not touch on pharmaceuticals, noted Sergey Morozov
| |
The Technical Committee "Artificial Intelligence" was created in 2019 at the initiative of RVC with the support Ministry of Industry and Trade RUSSIAN FEDERATION of Rosstandart. Standardization work in the field of AI in key industries economies will be carried out by TK 164 in accordance with the National Strategy for the Development of Artificial Intelligence for the Period up to 2030. It provides for the creation of unified standardization systems and conformity assessment of technological solutions developed on the basis of artificial intelligence, the development of international cooperation on standardization issues and ensuring the possibility of certification of products created on the basis of AI.
Sechenov University Joins International AI Standards Team in Medicine
Sechenov University takes an active part in the process of digitalization and informatization of healthcare. The next stage in the development of this area was the work within the framework of the working group on Artificial Intelligence TC 215 ISO Health Informatics of the International Organization for Standardization (ISO). Sechenov University announced this on February 18, 2020. Read more here.
| The site content is translated by machine translation software powered by PROMT. The machine-translated articles are not always perfect and may contain errors in vocabulary, syntax or grammar. Read original article If you find inaccuracies or errors in the results of machine translation, please write to editor@tadviser.ru. We will make every effort to correct them as soon as possible. |