Analysis of blood-brain barrier permeability based on machine learning approach.
Nadezhda Semenova1, Konstantin Sergeev1, Andrei Slepnev1, Oxana Semyachkina-Glushkovskaya1,2, Jürgen Kurths1,2,3
1Saratov State University, Russia
2Physics Department, Humboldt University, Berlin, Germany
3Potsdam Institute for Climate Impact Research, Potsdam, Germany
Abstract
The blood-brain barrier plays a decisive role in protecting the brain from toxins and pathogens. The ability to analyze the BBB opening (OBBB) is crucial for the treatment of many brain diseases, but it is very difficult to noninvasively monitor OBBB. In this work, we analyze the EEG series of healthy rats in free behaviour and after music-induced OBBB to estimate the permeability of this barrier on the basis of machine learning approach. For this purpose, we use a deep neural network with two hidden layers, which is trained on the realizations with OBBB and natural behaviour recorded for one group of animals; and then we apply the trained network to the realizations of another animals which are not involved in the training process. After the music impact the number of recognized OBBB is increased in about 50%. The proposed method is in a good agreement with other methods of EEG recognition.
This study is supported by RF Government Grant No. 075-15-2019-1885.
Speaker
Nadezhda Semenova
Saratov State University
Russia
Discussion
Ask question
