Yandex B2B Tech, the School of Data Analysis, and St. Petersburg State Pediatric Medical University (SPbGMU) have created a neural network that helps assess infant brain development. As emphasized in the Yandex report, this is the world’s first AI solution for patients of this age.
For diagnosis, it is necessary to download the patient’s MRI to a free cloud service. Patient information, including names and confidential details, is anonymized to protect privacy. The system requires only a few minutes to analyze the data and generate an image with outlined contours, indicating the percentage of gray and white matter in the child’s brain with an accuracy of more than 90%.
The report states that in the future, this solution will enable dynamic assessment of brain development and significantly speed up the delivery of research results to facilitate timely therapeutic decisions. This is particularly crucial in cases of suspected central nervous system pathologies, including cerebral palsy. Early detection of such conditions within the first months of life can enhance the effectiveness of treatment and improve the patient’s recovery prognosis.
The technology was tested at St. Petersburg State Pediatric Medical University. Professor Aleksandr Pozdnyakov, head of the Department of Medical Biophysics, emphasized that analyzing MRI images of infants and preparing a medical report can take from several hours to several days, whereas a neural network can substantially accelerate the diagnostic process and treatment selection.
Previously, Russia developed an IT solution capable of detecting severe congenital central nervous system diseases in fetuses during ultrasound examinations of pregnant women.