MosMedII Mosmed.ai

The main articles are:

• Artificial Intelligence in Medicine
• Artificial intelligence in radiology

2024

A new AI service for the diagnosis of cholelithiasis on computed tomography of the abdominal organs has begun to work in Moscow

In the capital, for the first time, an artificial intelligence service was launched to detect cholelithiasis on computed tomography. The neural network will automate the detection of gallstones, measure their size and speed up the diagnosis of pathology. This will help increase the chances of successful treatment and recovery of patients. Anastasia Rakova, Deputy Mayor of Moscow for Social Development, spoke about this on November 1, 2024.

The capital continues to actively introduce artificial intelligence services into the Moscow health care system. An algorithm has begun to detect gallstone disease on computed tomography of the abdominal organs. It is a common disease and is diagnosed in an average of one in five adult patients. Cholelithiasis can be asymptomatic for a long time without affecting the general condition of the patient. Therefore, computed tomography is considered one of the most reliable diagnostic methods in situations where the diagnosis remains unclear or requires additional verification. Computer vision technologies here act as an indispensable assistant radiologist: they pay the doctor's attention to the presence of gallstones, automatically make the necessary measurements, thereby speeding up diagnostics, - said Anastasia Rakova.

Before starting work with real patient research, the artificial intelligence service was tested to identify gallstones. This means that when working correctly, the algorithm does not recognize other changes as a deviation. The application of computer vision technologies in medicine helps to reduce the time for diagnosis and increase its accuracy.

Working on the introduction of artificial intelligence services in healthcare, we try to automate processes so that the doctor does not waste time on routine measurements. In addition, the use of such algorithms can serve as an additional means to verify the opinion of the doctor, allowing him to confirm his assessment in some cases. We share our metropolitan developments with the regions of our country: at the beginning of 2024, we opened access to doctors to the Moscow platform with artificial intelligence services "MosMedII." As of November 2024, 17 artificial intelligence services are available on the platform, which help doctors find signs of osteoporosis, breast cancer, pneumonia and other diseases, - said Yuri Vasiliev, chief radiologist of Moscow, director of the Center for Diagnostics and Telemedicine of the Moscow Department of Health.

Four more regions connected to the Moscow platform with AI services

Four more regions joined the Moscow platform "MosMedII." These are Kaliningrad, Lipetsk and Chelyabinsk regions, as well as the Yamalo-Nenets Autonomous Okrug. Now doctors of six constituent entities of the Russian Federation can use artificial intelligence (AI) services to process radiation studies. This was reported on August 13, 2024 at the Center for Diagnostics and Telemedicine with reference to the words of Anastasia Rakova, Deputy Mayor of Moscow for Social Development.

𝓲

NPCC DiT DBM

"The capital has been working for several years to automate routine processes, creating digital assistants for specialists from different fields. This allows you to better and faster provide medical care and, accordingly, achieve an increase in the duration and quality of life of Muscovites. At the beginning of 2024, we decided to share computer vision technologies and opened up access to doctors from the regions of Russia to a Moscow platform with artificial intelligence services for the rapid interpretation of computer tomograms, X-rays, fluorograms and mammograms. Medical institutions in the Voronezh and Moscow regions have been using smart algorithms for several months. Now they are also joined by the Kaliningrad, Lipetsk and Chelyabinsk regions, as well as the Yamalo-Nenets Autonomous Okrug. Thus, already six regions of the country have completed the connection to the capital platform. Our AI services analyzed over 160 thousand studies from these subjects, "said Anastasia Rakova.

17 artificial intelligence services are available on the MosMedII platform. They help doctors find signs of osteoporosis, breast cancer, pneumonia and other diseases. Neural networks mark areas of possible pathologies on the medical image with color prompts, and also automatically make measurements for diagnosis and compile descriptions of studies. Monitors the safety and quality of smart algorithms The Center for Diagnostics and Telemedicine of the Moscow Department of Health.

"Since 2020, we have been working with developers of artificial intelligence services. These are mainly domestic solutions. Our main difference from other countries - we know exactly for what purpose we use neural networks, we get obvious benefits for the doctor and health care in general. In addition, as a result of such work, 11 national standards and three pre-standards in the field of artificial intelligence in radiation diagnostics came into force. These are original developments that are not in any other country in the world, "said Yuri Vasiliev, chief radiologist of Moscow, director of the Moscow Center for Diagnostics and Telemedicine.

As a result of an experiment on the use of computer vision in radiation diagnostics, specialists developed and implemented a unique scientific methodology, on the basis of which more than 300 reference data sets were prepared. In addition, the first official library of AI services for healthcare in Russia was created, as well as a maturity matrix - an interactive panel that helps to make management decisions when using smart algorithms. For the convenience of doctors, a textbook on the preparation of data sets for neural networks in medicine has been published.

11 regions of the Russian Federation connected to the Moscow digital platform "MosMedII" for processing CT and X-ray images

In July 2024, 11 regions of Russia began connecting and testing the Moscow digital platform MosMedII, designed to process computer tomograms and X-rays using artificial intelligence (AI).

According to the Ministry of Health of Russia, the MosMedII platform provides regions with free centralized access to AI services capable of analyzing the results of computed tomography, radiography and mammography. Deputy Minister of Health of the Russian Federation Pavel Pugachev noted that Moscow, which has an experimental legal regime since 2020, acts as a supplier and experimental platform for creating innovative software products in the healthcare system.

𝓲

Unsplash

In the first two weeks of the platform's operation in pilot mode, more than 47.8 thousand radiation studies were processed using artificial intelligence technologies.

The introduction of AI technologies in medicine is accompanied by careful compliance with ethical standards. The Ministry of Health of Russia is guided by the Code of Ethics in the field of AI, which is designed to ensure the trust of doctors and patients in new technologies. The key principle remains that the final decision is always made by the doctor, and AI acts as an assistant who draws the attention of the specialist to certain aspects.

Experts of the Ministry of Health highlight a number of factors contributing to the development of AI in medicine: transparency of processes, consent and participation of the patient, responsibility of developers and operators, continuous training of systems and protection of confidentiality of medical data.

AI technologies in healthcare are already showing impressive results. So, when analyzing medical images, they are able to detect pathologies with an accuracy of 99.98%. At the moment, AI can identify 37 groups of pathologies out of the 120 most significant.

For further development and training of AI models, the Moscow Unified Medical Information and Analytical System (EMIAS) is ready to provide scientists and developers with marked up data.^[1]

Launch of 7 open datacets for mammography, chest X-ray, fluorography and computed tomography

Moscow has released 7 open datasets on mammography, chest X-ray, fluorography and computed tomography. Metropolitan Mayor Sergei Sobyanin wrote about this in his Telegram channel on May 3, 2024.

According to Sobyanin, a total of 68 open data sets have been created, they are used to assess and test the work of neural networks. Any developer can check neural networks for signs of pathologies of the brain, chest organs and other anatomical areas, the mayor noted.

𝓲

Freepik

Moscow has been engaged in the digitalization of the health care system for more than 10 years. Smart technologies are used daily and help make life easier for both doctors and patients, the mayor emphasized.

Neural networks are used in Moscow as part of an experiment to introduce computer vision technologies. The project implements a complex of social development in Moscow on the basis of the Center for Diagnostics and Telemedicine of the DZM with the support of the Department of Information Technologies.

Yuri Vasiliev, chief freelance specialist in radiation and instrumental diagnostics of the Moscow Department of Health, director of the Center for Diagnostics and Telemedicine of the DZM, believes that such algorithms are useful, "since approaches to patient therapy mainly depend not only on the fact of hemorrhage, but also on its location and volume." He stressed that for effective AI testing in this area, not only images confirming or refuting the very fact of hemorrhage are important, but also additional parameters included by Moscow scientists in the data set.

All open datacets are published on the platform mosmed.ai. They can be used in the self-testing of neural networks to search for signs of pathologies of the brain, chest organs and other anatomical areas.^[2]

In Moscow, created the first open set of data for AI training to identify brain pathology

Specialists from the Center for Diagnostics and Telemedicine of the Moscow Department of Health have created the first open set of data for training artificial intelligence (AI) to identify brain pathology. Such information was published on February 16, 2024 on the website of the capital's mayor's office.

A few years ago, Moscow, in fact, created a market for artificial intelligence in healthcare. Now we are increasing the pace of creating unique IT products for medicine, "says Anastasia Rakova, deputy mayor of the capital for social development. - One of the directions is the development by metropolitan scientists of impersonal datasets for assessing and testing neural networks. The last of them was a new unique dataset for intracranial hemorrhages.

𝓲

Mos.ru

According to the deputy mayor, the data bank includes about 800 CT studies, additional clinical and technical information. With its help, developers of artificial intelligence will be able to test neural networks to determine brain diseases, including the definition of intracranial hemorrhage, Rakova said. Datacets can also be used in the self-testing of neural networks to search for signs of pathologies of the brain, chest organs and other anatomical areas. All open datacets are published on the platform mosmed.ai.

Data Base contains sets of images in DICOM format and tables with markup, it is also supplemented with text conclusions of radiologists and information on types of hemorrhages, pathologies, technical characteristics.

According to Anastasia Rakova, by mid-February 2024, metropolitan specialists created only more than 350 datacets, among them 68 in the public domain. This helps in the development of high-quality artificial intelligence services throughout the country, she stressed.^[3]

Opening free access to medical AI services

On February 15, 2024, Moscow opened free access to its AI developments in the field for the regions. health care Details about this told the metropolitan mayor. Sergei Sobyanin

On behalf of President Vladimir Vladimirovich Putin, they decided that Moscow would share computer vision technologies and would open access to the best AI services in the capital for all medical organizations and regions of the country for free, Sobyanin said.

𝓲

Mos.ru

As part of the implementation of this project, the Center for Diagnostics and Telemedicine of the Moscow Department of Health has created a special platform "MosMedII," which regional health care institutions can use for remote analysis of radiation research.

Artificial intelligence helps specialists identify pathologies faster and make an accurate diagnosis. New equipment for diagnostics is connected to a single radiological center. The decryption results are uploaded to the electronic medical record.

To use it, you need to apply on the same site. Applications are accepted from medical institutions that have a valid license to carry out medical activities. After that, a bilateral agreement is concluded, which gives access to the capabilities of metropolitan medicine.

All images and research data should be anonymous, so it is impossible to "leak" the patient's personal data. The downloaded information is analyzed by artificial intelligence (AI). It is assumed that he will be able to direct the answer for an average of 15 minutes at any time of the day and night all year round.

As Sergei Sobyanin added, the use of AI algorithms created and used in Moscow in the regions will help to find signs of various diseases in medical images faster and more accurately.^[4]

2023

Moscow scientists presented an assessment of the maturity of AI services in radiation diagnostics

Moscow scientists assessed the maturity of artificial intelligence services operating in the capital's healthcare at the end of 2023. For this, a maturity matrix was used, previously developed by specialists from the Center for Diagnostics and Telemedicine of the city Department of Health. Using the service, radiologists can choose proven and reliable solutions based on artificial intelligence (AI). The NPCC DiT DZM announced this on May 14, 2024.

𝓲

NPCC DiT DBM

We strive to provide doctors with tools that not only simplify their work, but also improve the quality medical of care for Muscovites. Therefore, in 2023, we created a maturity matrix - a tool that shows the results of services and allows us to evaluate them. At the end of 2023, we note an increase in technical stability and diagnostic accuracy. neuronets The maturity matrix showed that the best results for 2023 are demonstrated by services for assessing computer and tomographic signs of various pathologies of the abdominal organs, signs of, breast cancer to data mammographies radiographic signs of various pathologies of the musculoskeletal system, as well as comprehensive services for assessing CT signs of various pathologies of the chest organs. Thanks to this tool, doctors can choose the highest quality and most reliable solutions from existing ones for their work, - said the deputy head of the Moscow City Health Department. Ilya Tyrov

The maturity matrix allows medical institutions to choose the best AI services for themselves, and developers to assess the dynamics of their development and competitiveness. Neural networks used by Moscow doctors help radiologists find signs of various diseases on medical images.

The methodology for assessing the maturity of artificial intelligence services was compiled on the basis of monitoring the quality of software products during the Moscow experiment on the use of computer vision technologies in healthcare. This is the largest scientific study carried out in the capital for the fifth year.

When implementing artificial intelligence in healthcare, it is important to opt for mature solutions. This is technically stable software that has high diagnostic accuracy metrics. The success factors in this case can be considered the maturity of the development companies and the scientific and technical support of the experts of our center who oversee the experiment. We place great emphasis on the safety of those solutions that are allowed to work with the data of the capital's patients: we select and control the work of all solutions based on artificial intelligence, "said Yuri Vasiliev, chief radiologist of Moscow, director of the Center for Diagnostics and Telemedicine of the Moscow Department of Health.

In Moscow, the official library of data sets for the healthcare sector has been replenished by 50%

In Moscow, the official library of data sets for the healthcare sector has been replenished by 50% per year.

Ilya Tyrov, deputy head of the Moscow City Health Department, spoke about this on August 21, 2023.

{{quote 'The capital in every possible way supports the development of artificial intelligence in healthcare. This is also evidenced by the pace of introduction of neural networks in this area and the creation of IT products for medicine. Thus, the official library of data sets for the assessment and training of artificial intelligence for the year increased by 50 percent: as of August 2023, it has more than 300 datacets. They help developers in testing and further training their own intellectual solutions for the analysis of radiation research as part of the Moscow experiment on the introduction of computer vision in medicine, "said Ilya Tyrov. }}

The purpose of published data sets is different: for conducting a technical or diagnostic self-test of a neural network, scientific research, calibration or functional testing of an artificial intelligence service as part of a Moscow experiment on the introduction of computer vision technologies into medicine. The kits are prepared by specialists from the Center for Diagnostics and Telemedicine of the Moscow Department of Health. The library appeared in 2022 and is replenished monthly.

As of August 2023, in more than 20 areas, the developers went to streaming research, testing and refinement of models are being carried out for the rest. It is important that it is conducted on the basis of a stream of real impersonal studies of patients and doctors constantly provide feedback on the work of algorithms. Equal conditions have also been created for all participants: the catalog of solutions based on artificial intelligence and the matrix of technological and clinical maturity of services are constantly being updated, "said Yuri Vasiliev, director of the Center for Diagnostics and Telemedicine.

One of two doctors analyzing mammographic images in Moscow replaced by artificial intelligence

One of the two doctors analyzing mammographic images in Moscow has been replaced by artificial intelligence. A similar practice is planned to be extended to all of Russia. This was announced at the end of July 2023 by Deputy Minister of Health Pavel Pugachev.

According to our requirements, in the process of analyzing the results of mammography, there should be two readings by two different medical professionals. Moscow was the first to propose using an AI program instead of a second reading. First, he "looks," makes a preliminary conclusion, after which the doctor validates it, agrees with the opinion of AI or not. We have studied this experience and are working on the issue so that it can be regulated throughout the country and abandoned the second reading, replacing it with AI, "the deputy minister explained.

Russian neural network "sees" breast cancer on mammography as a real doctor

As a result of the use of AI, Moscow radiologists have accelerated the process of analyzing images, they "read" the results of mammography eight times faster than before. At the same time, the quality of diagnostics did not get worse.

Since the beginning of 2023, doctors have been analyzing mammographic studies of patients in Moscow using artificial intelligence within the framework of compulsory health insurance (compulsory medical insurance ). During this time, neural networks have already analyzed more than 100 thousand such images, said Anastasia Rakova, Deputy Mayor of Moscow for Social Development, at the end of June 2023 .

Smart algorithms are able to describe a large number of medical images faster than humans. Research by Moscow scientists has proven that artificial intelligence services reduce the time for describing mammograms by more than eight times, while maintaining the high quality of diagnostics. Thus, patients earlier receive research results, and doctors save significant temporary resources, she said.

The innovation was implemented as part of a Moscow experiment on the use of computer vision technologies in medicine, which has been successfully carried out for the fourth year on the basis of the Center for Diagnostics and Telemedicine by the Moscow Department of Health and the Department of Information Technologies. By mid-2023, about 50 algorithms are processing research in 21 clinical areas.^[5]

2022

In Moscow, 50% of radiation studies of polyclinic patients are processed by artificial intelligence

In Moscow, neural networks have been helping radiologists for more than two years. Their capabilities are regularly expanded and the proportion of images that are processed by smart algorithms is constantly growing. At the beginning of December 2022, already half of all radiation studies of patients in Moscow polyclinics analyze artificial intelligence services. This was announced on December 1, 2022 by the deputy head of the Health Department of the city Ilya Tyrov.

Neural networks are actively developing and are increasingly used in Moscow healthcare. If two years ago they worked only in one direction of radiation research, then as of December 2022 there are already 17. Also, all the digital equipment of polyclinics was combined into a single network, and now the pictures are uploaded to the digital cloud - a single radiological information service of a single medical information and analytical system. Thanks to this, the share of studies processed by smart algorithms has grown significantly, - said Ilya Tyrov.

As of December 1, 2022 algorithms , 150 medical institutions, including children's, are available to radiologists. In total, more than 40 artificial intelligence services work in 17 different areas of research. Doctors use neural networks for processing fluorography, mammography, radiography, and. computed tomography magnetic resonance imaging Artificial intelligence helps to find on radiation studies signs,, lung cancer COVID-19 osteoporosis of the spine, aorta, aneurysms ischemic disease, hearts stroke pulmonary hypertension, hydrothorax, as well as,, breast cancer hernias flat feet and other diseases. In addition, in the arsenal of doctors - a comprehensive service, which on one image medical of tomography computer finds signs of seven pathologies at once.

Every month we compile a rating on performance indicators. It is available on the project website mosmed.ai. Artificial intelligence is in the process of development, but we have achieved that computer vision technologies have significantly increased their performance. The use of algorithms allows you to reduce the load, acts as an additional assistant to radiologists, - said Yuri Vasiliev, director of the Moscow Center for Diagnostics and Telemedicine of the DZM, chief freelance specialist in radiation and instrumental diagnostics of the DZM.

The algorithms work as part of an experiment to introduce computer vision technologies into healthcare - a project of the Social Development Complex of the Moscow City Hall on the basis of the Center for Diagnostics and Telemedicine of the DZM with the support of the Department of Information Technologies.

Adding Basic Radiological Diagnostic Modalities

On October 18, 2022, the Center for Diagnostics and Telemedicine DZM informed Zdrav.Expert that a digital library of impersonal data sets for assessing and training neural networks was expanded in Moscow.

Moscow has opened a digital library for neural networks in the main modalities of radiation diagnostics
Photo source: arscomp.ru

Now it is presented in all the main modalities of radiation diagnostics, as well as in ECG and ultrasound. Thus, developers of artificial intelligence services from any region can already access 40 datacets. This will help in testing and further training your own intelligent solutions for the analysis of medical research.

Moscow radiation diagnostics has been using artificial intelligence services for several years. They facilitate the work of metropolitan doctors and help not to miss even the most invisible signs of pathology. We have accumulated extensive experience in this area, and we are ready to share our work. To do this, we opened access to a library of datasets, where anyone can already access 40 datacets for most of the main modalities, for example, ultrasound, electrocardiography, radiography, computer and magnetoresonance tomography and others. Developers, researchers, students and other users can freely download such datasets according to the desired modality and anatomical area to test their own neural networks, told Ilya Tyrov, Deputy Head of the Moscow City Health Department.

Datasets are sets of impersonal radiological images created as part of the Moscow experiment on the introduction of computer vision into medicine. The library of open datacets appeared three months ago, today the number of sets in it has more than 4 times increased - to 40. All of them are published on the platform mosmed.ai

mosmed.ai

Data sets can be used in self-testing of neural networks to search for signs of pathologies of the cardiovascular system, brain, chest organs and other pathologies. The most popular is the CT image of the chest organs, in order to train AI services to search for signs of COVID-19, it was viewed more than 4 thousand times and downloaded by over 700 users. In second place is the combined data set for several types of self-testing studies at once.

The second place is taken by the combined dataset for several types of research at once, such as computed tomography, mammography, X-ray, fluorography. Closes the top three brain MRI data set.

The purpose of published data sets is different: for conducting a technical or diagnostic self-test of a neural network, scientific research, calibration or functional testing of an artificial intelligence service within the framework of the Moscow experiment. After successful self-testing, the developer can apply for participation in the Moscow experiment, and if the stages of the experiment are successful, the algorithm can be introduced into the work of the capital's healthcare. In this regard, our Center is different: we provide the entire cycle of services related to the creation and operation of algorithms - from assistance with determining the direction of the neural network and its testing to clinical trials and implementation in clinical practice. noted Yuri Vasiliev, chief freelance specialist in radiation and instrumental diagnostics of the Moscow Department of Health, director of the Center for Diagnostics and Telemedicine of DZM.

The experiment on the introduction of computer vision into medicine was launched by the Moscow Social Development Complex on the basis of the Moscow Center for Diagnostics and Telemedicine and the city Department of Information Technologies. The center has become a platform for the development of artificial intelligence technologies in Russia. Artificial intelligence services have been integrated into the Unified Radiological Information Service of the Unified Medical Information and Analytical System (EMIAS). Thus, radiologists of all medical institutions in Moscow connected to EMIAS were able to use innovative technologies.

Moscow has been actively working on the digitalization of the health care system for ten years. For October 2022, the basis of this process is a single digital healthcare platform. It provides personalized management of each patient at all stages: from diagnosis and treatment to follow-up. Thanks to the platform, all data on the state of health of citizens are available online both to doctors and to patients themselves.

Links

Platform Link

Notes