Cancer diagnostics in Tomsk Oncology Center will be carried out using a neural network
Every day, artificial intelligence identifies an average of seven to eight patients who should be sent for additional examination
TOMSK, November 9. / TASS /. Artificial intelligence, developed by the Sberbank laboratory, will be used by doctors at the Tomsk Regional Oncology Center for the analysis and processing of mammographic images. This digital product will reduce the time spent on diagnosing breast cancer and, accordingly, prescribe treatment to patients faster, the press service of the regional administration told reporters.
"The laboratory for artificial intelligence (AI) of Sberra and the Tomsk regional oncological dispensary have introduced an AI model for the analysis and processing of mammographic images. Today, the process of analysis of mammograms with the help of AI in the Tomsk Oncology Center is fully automated and synchronized with the diagnostic devices of the medical institution. The underlying AI model is The neural network evaluates images literally "on the fly": the image is processed within a few seconds, and in the event of a probable pathology, the doctor receives the image's assessment from the system and examines and describes it more thoroughly, "the message says.
As the press service explains, the main advantage of the model is the ability to quickly process a large number of images. “This saves time not only for doctors, but also for the patients themselves - they begin to receive the necessary treatment faster, which ultimately preserves health and saves lives. Together with doctors, we see great prospects for this product and plan to introduce it in other Russian regions. ", - quoted in the message by the first deputy chairman of the board of Sberbank Alexander Vedyakhin.
It is clarified that this diagnostic method has already been tested at the dispensary, it lasted several months. "The test showed that the machine evaluates the images no worse than doctors. Just a few seconds after processing, the doctor receives an encoded evaluation of the image from the system and takes it into account as a second opinion. Every day, artificial intelligence identifies an average of seven to eight patients who should be sent to additional examination ", - the words of the head of the department of X-ray diagnostics of the oncologic dispensary Yevgeny Karpov are quoted in the message.
It is also reported that the neural network is constantly evolving and, according to experts' forecasts, will reach the level of the most high-class diagnosticians in a year.
About the project in the Tomsk region
Digitalization of medicine is one of the priority areas of the national project "Health", which is being implemented in the Russian Federation by decree of Russian President Vladimir Putin. The national project involves reducing the mortality rate of the working-age population (up to 350 cases per 100 thousand of the population) until 2024, eliminating the personnel shortage in medical organizations providing primary health care, ensuring that all citizens are covered with preventive medical examinations at least once a year. In total, it is planned to spend more than 1.7 trillion rubles on the implementation of the national project in the Russian Federation in 2019-2024.
According to the regional administration, within the framework of the federal project "Fight against oncological diseases" (implemented as part of the national project "Health") in the region by 2024 it is planned to reduce mortality from neoplasms from 218.1 cases per 100 thousand people to 205 cases. Early-stage cancer detection is expected to rise from 57.2% to 63%. For the implementation of the federal project in 2019, the region will receive about 2.3 billion rubles.