Care Mentor AI the Neuronet for the description of x-rays

Main articles:

• Neuronets (neural networks)
• Artificial intelligence in medicine
• Artificial intelligence in radiology

Service based on a medical neuronet for the analysis of radiological researches (KT, MPT, X-ray, etc.)

2021: Plans of integration of the neural network Care Mentor AI system into Sbermediya's ecosystem

"Sbermediya"and Russian the developer neural network of platforms of Ltd company "КэременторЭйАй" AI (Care Mentor) agreed about start of joint projects for digitalization Russian health care. medical Russia In 2021 in clinics the advanced software products for beam diagnostics based on will be already implemented artificial intelligence (AI). On January 29, 2021 reported about it century Sberbank.

The ecosystem for development of AI in health care of Russia — Sbermediya — not only offers own services in health care, but also integrates developments of the Russian companies. The first participant of public market who joined an ecosystem — Kerementoreyay LLC company (Care Mentor AI). The developed neural network Care Mentor AI system for determination of pathologies of bodies of a thorax will be integrated into Sbermediya's ecosystem and in the nearest future will become available to users. It is expected that integration of the advanced services at a single platform will allow to construct and implement effective support systems of adoption of medical solutions and also will give new opportunities for scaling of digital services of developer companies across all territory of Russia.

The system of computer vision analyzes the medical beam image, determines probabilities of pathologies and performs prioritization of cases with pathological finds for doctors. Such mode allows health workers to concentrate quicker on complex cases in radiological researches, to reduce time for adoption of the final diagnostic solution and also to increase personal efficiency and performance.

The Care Mentor AI system is already implemented in a number of regions of Russia, and in the long term can be applied to automation of diagnostic process in scales of department, all medical organization and also in general to a didzhitalization of delivery of health care in Russia. In plans of partners — cooperation in digitalization of programs of medical examination of Russian regions. And by the end of 2021 partner integration on all line of services of the Care Mentor AI platform is supposed.

On the platform Sbermediya the best developments in the field of AI for health care integrate. Cooperation with Kerementoreyay LLC (Care Mentor AI) will allow to propose to clients the qualitative and effective solution helping the doctor in the analysis of medical beam images. It is sure that the product will be extremely demanded by professional community — Ekaterina Kolomentseva, the managing director of the Center of medical products and services Sbera noted.

For us as system integrator of the advanced solutions AI for health care, agreement signature with Kerementoreyay LLC company whose developments already won location of medical community is especially important and are widely applied, in particular, within the Moscow experiment on implementation of AI in health care. State registration of software as medical product brought closer implementation of new technology in clinical practice for all Russia in what, undoubtedly, connection to our ecosystem will help. It is open for joining of other market participants whom we invite to cooperation — Igal Zuk, the CEO of the company of Sbermediya reported.

We are glad to begin work with such systemically important enterprise as Sbermediya that together more effectively and quickly to provide the state hospitals of regions of Russia with advanced technologies of computer vision. Neural network technologies will facilitate diagnosing process, having acted as the electronic medical assistant. We thank Sbermediya's company for the put trust to our product and we hope that our joint efforts will bring the Russian health care to qualitatively new level — Ilya Plisko, the CEO of Kerementoreyay LLC company (Care Mentor AI) said.

2020

Registration in Roszdravnadzor as a medical product

The resident Skolkovo, Care Mentor AI company announced on December 21, 2020 registration in Roszdravnadzor as a medical product of the software "A system neural network Care Mentor AI" (the registration certificate of RZN 2020/11137).

The system of computer vision analyzes the medical beam image, determines probabilities of pathologies and visualizes finds. Such mode allows doctors to concentrate quicker on "interest zones" in radiological researches, to reduce time for adoption of the final diagnostic solution, and, as a result, to increase personal efficiency and performance. The Care Mentor AI system can be applied to automation of diagnostic process in scales of department and all medical organization.

For us obtaining the registration certificate from the first and without failures — the sign of great trust from the state to AI services generally and to our system in particular. Now we can fully work with the medical organizations, the diagnostic centers, hospitals and clinics and to help doctors in all territory of the Russian Federation, removing from them a part of routine work. Our service automatically makes the work sheet of the doctor with prediction of pathologies.

The solvency of a system from the technical and functional points of view was proved during the experiment on use of the innovative technologies in the field of computer vision for the analysis of medical images, further application in a health care system of Moscow and also during technical and clinical trials. After obtaining the registration certificate, Care Mentor AI it is ready to scaling of the technologies including at the federal level within accomplishment of the national project on digitalization of health care.

Obtaining the registration certificate of Roszdravnadzor - serious break for Care Mentor AI company and one more significant step on the way to digitalization of the Russian medicine using artificial intelligence. Service of the resident of Fund will become the irreplaceable assistant to radiologists that is especially important during a pandemic.

Connection to Uniform radiological information service for the analysis of KT-pictures

On October 27, 2020 the company Care Mentor AI announced that of the same name AI- the service analyzing KT-researches and revealing on them symptoms koronavirusny of an infection was connected to To uniform radiological information service EMIAS (ERIS). The Care Mentor AI system defines percent and extent of damage of lungs at patients with COVID-19 and issues visualization finds on series KT- lungs.

Integration takes place within the large-scale Moscow experiment started Moscow Department of Healthcare and Center of diagnostics and telemedicine since February, 2020. Its purpose – comprehensively to investigate opportunities software (SOFTWARE), created on the basis of technologies computer vision, during the work in medical the organizations of the capital.

"When at the beginning of a year the pandemic of a koronavirusny infection began, we were already ready to connection of services on the basis of artificial intelligence in Uniform radiological information service. Algorithms to the doctor automatically range researches that doctors first of all described researches with a high probability of viral pneumonia. After that algorithms automatically, on developed by us by methodology, create templates of descriptions which radiologists can turn on in the text of the conclusion. It helps doctors" — Sergey Morozov, the chief specialist on beam and instrumental diagnostics of Moscow, the director of the Center of diagnostics and telemedicine emphasized.

"Thanks to work with the medical organizations of Moscow, we really help a large number of both doctors-diagnosticians, and patients. Our command is grateful to organizers of an experiment for the given opportunity to us and to other developers quickly to implement the projects in Health care of Moscow. In a pandemic all medical sector faced a set of problems on which solution there was critically not enough time, but we threw all forces on solution development which helps doctors-diagnosticians to define COVID-19 on KT-researches, and succeeded in it" — Ilya Plisko, the CEO of Care Mentor AI said.

Involvement of the Zhores supercomputer for increase in accuracy of determination of extent of defeat of COVID-19

Within the program for fight against COVID-19 of the Center of Skoltech for scientific and engineering computing technologies for tasks with data bulks (CDISE) date scientists of Care Mentor AI company could use the Zhores supercomputer for increase in accuracy of determination of pathologies. On September 2, 2020 the Care Mentor AI company reported about it. Read more here.

Integration into Uniform radiological information service for the analysis of roentgenograms

The software of Care Mentor AI which analyzes was integrated into the Uniform Radiological Information Service (URIS) to which the diagnostic equipment of Moscow is connected and carries out screening of roentgenograms of bodies of a thorax on presence of different pathologies, including such socially important as lung cancer, tuberculosis, pneumonia. This Zdrav.Expert became known on July 3, 2020. Care Mentor AI will allow to analyze radiographic images and within several seconds to determine existence or lack of pathological changes of bodies of a thorax by eight groups of signs (including socially important diseases). Read more here.

Development of AI for operational diagnosing of COVID-19 according to roentgenograms and KT of patients

On April 8, 2020 the Care Mentor AI company reported that it develops the system of Artificial Intelligence which will accelerate the analysis of survey roentgenograms of a thorax and KT of patients with suspicion of viral pneumonia which for April, 2020, taking into account an epidemiological situation, is in most cases caused by a coronavirus. For April, 2020 the moment neuronet accuracy on radiological researches is 84% (on a part of internal data). The company works with public data sets, at computing powers, a part from which is provided by the partner, NVIDIA company. Data are marked by experienced radiologists in the interface which is specially developed by the company.

According to the company, the command works on researches in coauthorship with the leading Russian scientific and clinical collectives: during work the company involved experts in the field of radiodiagnosis of St. Petersburg State University and Mariinsky hospital. The company created the innovation systems of computer vision on the basis of artificial neural networks. These systems it is used for detection of pathologies of lungs according to roentgenography and also it will be used for detection of pathologies of lungs according to native KT of a thorax. Neural network models do not concede on accuracy to the qualified radiologist, however exceed human capacities on productivity.

The company offers the first to use a system for the analysis of survey roentgenograms of a thorax. In modern conditions of universal availability of the x-ray equipment and relative low cost of a method, it can cardinally change in a short space of time a situation with identification of cases in the country.

2019

Participation of SPb NIIF in development and testing of a neuronet of Care Mentor AI

On November 15, 2019 TAdviser knew that specialists of SPb of scientific research institute of a ftiziopulmonologiya under the leadership of professor Pyotr Kazimirovich Yablonsky participate in creation of a neuronet of Care Mentor AI.

In particular, in October the director of Federal State Budgetary Institution SPb NIIF Russian Ministry of Health professor Pyotr Kazimirovich Yablonsky signed two agreements with Care Mentor AI company: the first — about testing of Care Mentor AI network according to the analysis of digital roentgenographies of lungs; the second — about participation of scientists of institute in development of an algorithm of training and an analysis system of images of a neuronet of Care Mentor AI — native KT of lungs.

For us cooperation with Institute of a ftiziopulmonologiya is extremely perspective and, at the same time, responsibly. Our St. Petersburg colleagues created and registered electronic databases of results of beam researches of lungs with the verified pathological changes of both tubercular, and other origin. Besides, scientists of institute already have rich experience of clinical approbation of different automated systems of the analysis of diagnostic images. Application of unique data sets for testing of artificial intelligence will allow to carry out independent examination of diagnostic accuracy of the neural network product developed by us at the highest scientific and methodical level — Dmitry Blinov, the head of department of Care Mentor AI of scientific work emphasized.

For us cooperation with specialists from the Russian company Care Mentor AI developing the systems of artificial intelligence for medicine, undoubtedly, interestingly. We understand that disruptive solutions in medicine are possible only on the basis of rational use of high technologies, such as analysis of Big Data, machine learning, etc. It is pleasant that developers of Care Mentor AI company understand that without close cooperation with leading specialists in the field of clinical medicine it is impossible to create really necessary, working software or technology product — Pyotr Kazimirovich Yablonsky, the director of Federal State Budgetary Institution SPb NIIF Russian Ministry of Health, MD, professor, the honored doctor of the Russian Federation, the chief non-staff specialist the thoracic surgeon of the Russian Ministry of Health noted.

According to the director of Federal State Budgetary Institution SPb NIIF, carrying out the joint project on development of the AI platform for the analysis of results of native KT of a thorax is planned. Successful project implementation will allow to reduce labor costs of radiologists on identification of symptoms of such socially important diseases as tumoral defeats and tuberculosis of lungs, when preserving the high level of diagnostic accuracy.

Pyotr Kazimirovich himself is going to act as the scientific project coordinator. Rather labor-intensive process is necessary to scientists and developers: specialists of NIIF from a team of professor Yablonsky should mark a large set of KT of pictures on which the network will be trained, then test stages, corrections, after-training of network, repeated testing, etc. are coming. Special attention in this work will be paid to investment of neural network with capability to reveal small, up to 5-10 mm in size, focal changes in lungs. The first project deliverables are supposed to be received in the spring of 2020.

Availability of solutions for radiologists of Care Mentor AI on a show-window of CoBrain

On October 22, 2019 the Skolkovo institute of science and technologies announced (Skoltech) that according to the signed agreement on a show-window of the CoBrain platform the solutions of a system of artificial intelligence for the analysis and interpretation of results of beam methods of the research Care Mentor AI analyzing roentgenographies of bodies of a thorax, cervical department of a backbone, foot and knee joints are presented. Read more here.