Rakova spoke about the creation of more than 350 datasets for training medical neural networks
[ad_1]
The first set of data on computed tomography of the brain was created by specialists from the Center for Diagnostics and Telemedicine of the Moscow Department of Health. It contains 800 CT examinations, as well as clinical and technical information. The data will help developers test neural networks that can detect signs of intracranial hemorrhage.
“Several years ago, Moscow essentially created the AI market in healthcare. Now we are increasing the pace of creating unique IT products for medicine. One of the directions is the development by metropolitan scientists of anonymized data sets for evaluating and testing neural networks,” said Deputy Mayor of Moscow for Social Development Anastasia Rakova.
The vice mayor added that the latest was a new unique data set on intracranial hemorrhage. It includes the results of 800 CT examinations and clinically important features. Taking them into account will help make the search for pathologies more effective.
“To date, our specialists have created more than 350 such datasets, including 68 in the public domain. This helps in the development of high-quality AI services throughout the country,” the deputy mayor noted.
The data set is necessary for companies creating tomogram analysis systems for determining skull fractures, hemorrhages, assessing the type by location, and their multiplicity. All open datasets are published on the mosmed.ai platform.
[ad_2]
Source link
