Simplified distributed dipole model establishment and identification method based on D-K partition

Abstract

The invention discloses a simplified distributed dipole model establishment and identification method based on D-K partition, and the method specifically comprises the steps: firstly, carrying out the filtering of an original MI-EEG through different band-pass filters, so as to select an optimal frequency band related to the motor imagery activity; then, electroencephalogram inverse transformation is carried out on each selected sub-band, and scalp EEG is converted into dipoles in the cerebral cortex; thirdly, obtaining a central dipole based on neuroanatomy D-K partition to construct a simplified distributed dipole model, regarding the activity of the cerebral cortex central dipole as a neurodynamic system, and constructing 4D data expression; and finally, fusing the multiband data expressions, inputting the fused multiband data expressions into a designed n-branch parallel nB3DCNN, and carrying out comprehensive feature extraction and recognition from three dimensions of time, frequency and space. According to the invention, under different frequency bands, the amplitude of the dipoles in the 3D space changes along with time, and a small number of central dipoles are utilized to reflect the neural electrical activity caused by motor imagery of the whole cerebral cortex layer.

Description

technical field [0001] The invention relates to brain source imaging (ESI) and three-dimensional convolutional neural network (3DCNN), which are used in the technical field of motor imagery electroencephalogram (MI-EEG) recognition. Background technique [0002] Brain-computer interface (BCI) technology based on motor imagery is an important direction in neurorehabilitation research. How to accurately detect changes in the cerebral cortex is crucial to the decoding of motor imagery. Through the establishment of a head model and EEG signals measured in the scalp, equivalent current dipoles can be calculated with high spatiotemporal resolution to represent neural electrical activity within the brain, providing new insights into cortical dynamics Research ideas. However, if the dipoles of the entire cortical layer are used to construct features, it may cause information redundancy, and the cost of generating feature data is too high. It will result in the loss of key informat...

AI Technical Summary

Problems solved by technology

However, if the dipoles of the entire cortical layer are used to construct features, it may cause information redundancy and the cost of generating feature data is too high; if the region of interest (ROI) is selected, only part of the dipoles are used to construct features, and it may be loss of key information

Images