NPCC DiT DZM: HUB Telemed Automated X-ray Analysis Service

The main articles are:

• Radiology
• Artificial intelligence in radiology
• Artificial Intelligence in Medicine

2023: Analysis of 10 million radiation studies

Neuronets Moscow to radiologists helped analyze 10 million radiation studies. algorithms On average, 270 thousand radiation studies are processed by smart specialists per month. This was announced on June 21, 2023. NPCC DiT DZM (Scientific and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Department of Health)

We are conducting systematic scientific and practical work to introduce artificial intelligence into healthcare. Specialists are working on the introduction of the best solutions in the field of medicine - they accompany the process and control the quality of neural networks. Radiologists of medical institutions in the capital, whose diagnostic devices are connected to the unified radiological information service EMIAS, can use more than 50 algorithms in 21 clinical areas to describe medical images, - said Anastasia Rakova.

Use of computer vision technologies in medicine makes it possible to increase efficiency of radiation diagnostics. In the image, algorithms mark areas of possible pathologies with color hints, take measurements that are of clinical value to the physician for diagnosis, and make a description. Computer vision technology makes it possible to find pathologies at an early stage, when they can be invisible to the human eye.

Artificial intelligence allows finding signs, lung cancer COVID-19 osteoporosis spine, aneurysms aorta, ischemic disease, hearts stroke pulmonary hypertension, hydrothorax, herniation of intervertebral breast cancer discs, flat-footedness and other diseases in images of radiation studies. There are also comprehensive artificial intelligence services that can detect medical up to 10 pathologies on a single image.

Neural networks process fluoro-, mammo- and radiography, computer and magnetic resonance imaging. In 2022, other regions of Russia joined the project using the platform of the Center for Diagnostics and Telemedicine of the Moscow Department of Health Hub Telemed.

Artificial intelligence has established itself as a reliable doctor's assistant and we are ready to connect the regions of Russia to our Experiment. Radiologists of the Yamal-Nenets Autonomous Okrug and the Vsevolozhsk Clinical Interdistrict Hospital of the Leningrad Region for June 2023 have access to 7 artificial intelligence services in 4 areas of research. In the Yamalo-Nenets Autonomous Okrug, more than 27 thousand medical images have already been processed by neural networks. Through the platform of the Center for Diagnostics and Telemedicine of the Metropolitan Department of Health HUB Telemed, studies of regional clinics are transmitted anonymously for analysis by artificial intelligence, after which they are returned to the doctor for a final description. Our specialists control this process, interact with developers, collect and analyze feedback from doctors and developers, this allows improving the quality of algorithms. We offer other regions of our country to join our project, - said Yuri Vasiliev, chief freelance specialist in radiation and instrumental diagnostics of the Moscow Department of Health, director of the Center for Diagnostics and Telemedicine.

2020: Description

As of 2020, Hub Telemed is designed to process computed tomography of the lungs of patients with suspected COVID-19, as well as to assess the dynamics of an already confirmed disease. The service also allows you to assess the results of mammography.

The service allows doctors from the regions to remotely receive the necessary expert advice, they can share their opinion in difficult cases or perform the initial description of the picture. The diagnostic time is less than 15 minutes.

With the help of Hub Telemed, you can become a participant in a federal project to compile a database of reference datacets, which assess the accuracy of medical and artistic intelligence. To do this, you need to download the data of the computed tomography study of a patient with confirmed COVID-19^[1].

Notes