The Moscow Scientific and Technical Center “Module” has developed artificial intelligence algorithms that recognise viral pneumonia, including one caused by COVID-19, with an accuracy of 93%. According to Aleksey Mokhnatkin, Deputy Director for Development of the STC "Module", the neural network can be easily integrated into the IT infrastructure of medical institutions. While she works as part of the demonstrator of the hardware and software complex.

A neural network with the U-Net architecture is used to recognise the affected areas of the lungs. It assumes the possibility of using the development in other medical fields and for the diagnosis of other diseases.

Original (in Russian)