Artificial intelligence has learned to look for impaired brain function

A group of Russian and German neurophysiologists presented artificial intelligence, which can analyze the signals of the electrical activity of the brain and find traces of various disorders in its work.

The development of artificial intelligence under the leadership of Alexander Khramov from the Laboratory of Neuroscience and Cognitive Technologies at Innopolis University allowed the neural network to search for deviations from the norm according to the data of the brain electroencephalogram.

The basis of artificial intelligence is a direct distribution neural network. This type of neural network allows you to create not multifunctional, but fast algorithms that specialize in one main task.

The neural network was taught to look for similar patterns in the signals of the electrical activity of the brain. As part of the study, scientists analyzed data from two brain regions in six rats predisposed to the development of epilepsy.

Artificial intelligence showed how the type of interactions between the cortex and the deep structures of the brain, presumably associated with the development of epilepsy, changes when another attack occurs. At the same time, his calculations allowed us to consider and compare the type of interactions between these areas of the nervous system during a normal state, as well as before the onset of a seizure.

In the future, these algorithms can be used to create large-scale artificial intelligence for the analysis of human brain data.

hightech.fm