Lobachevsky University researchers discovered how to stimulate hippocampal activity using a neural network
As part of their efforts to improve technologies for restoring brain function in neurodegenerative diseases, scientists at Lobachevsky University have developed a neurointerface for modelling and triggering the electrical activity of the hippocampus. By controlling artificial neurosignals, it is possible to stimulate the formation of memory and learning ability of the brain. This is Russia's first technology for restoring activity in the hippocampus. The technology will enable the development of neuroprosthetics to restore brain function in epilepsy, dementia and Alzheimer's disease.
"We conducted the experiment on an in vitro brain slice of laboratory animals. We recorded many signals in a healthy hippocampus, uploaded these data to a neural network and taught the AI to predict the electrical activity of the hippocampus. We were able to reproduce the signals modelled by the neural network in the damaged part of the brain and "start" the hippocampus that was not working," said Albina Lebedeva, Senior Researcher at the UNN Research Institute of Neurosciences.
According to the scientist, the system allows controlling the parameters of neurosignals, replicating the behaviour of a healthy brain as closely as possible. The software system consists of a neural network, custom software and a circuit board with electrodes for reading and reproducing neurosignals. The system allows converting digital neurosignals modelled by the AI into standard electrical activity of the brain.
"The technology is built on the arduino uno platform. We managed to adapt and improve the platform code to convert and transmit signals to the hippocampus. The software allows us to adjust the frequency, duration, amplitude and shape of the signal," said the author of the code, Ivan Kipelkin, Junior Researcher at the UNN Research Institute of Neurosciences.
Experts from the Technology Transfer Centre of Lobachevsky University have performed state registration of the programme and obtained an appropriate certificate.
"For the in vivo experiment, we are developing a hippocampus stimulation unit for laboratory animals equipped with a VR system: the animal will run on a ball, moving in a virtual maze. We will read and reproduce neurosignals in real time, which will allow us to fine-tune and improve the neurointerface," said Albina Lebedeva.
In the future, researchers intend to design a neurochip implant to restore or improve the functioning of the hippocampus. The neurointerface was developed by scientists from the Research Institute of Neurosciences and the Department of Control Theory and System Dynamics at the UNN Institute of Information Technology, Mathematics and Mechanics. The research is supported by a grant from the Russian Science Foundation and is performed within the framework of the state assignment by the Youth Research Laboratory for Artificial Intelligence and Big Data Processing.
The results have been published in the international scientific journal Algorithms.
