Epileptic electroencephalogram recognition method and system based on convolutional neural network

Abstract

The invention discloses an epileptic electroencephalogram recognition method and a recognition system based on a convolutional neural network, wherein the method comprises the following steps: acquiring electroencephalogram signals via an electrode, and analyzing and processing the electroencephalogram signals, so that electroencephalogram pictures are obtained; and inputting the electroencephalogram pictures into the trained convolutional neural network, so as to judge whether the electroencephalogram pictures have epileptic discharge labels or not; and the convolutional neural network is trained by a process comprising the following steps: collecting sample electroencephalogram pictures, wherein the sample electroencephalogram pictures include sample electroencephalogram pictures havingthe epileptic discharge labels and sample electroencephalogram pictures having no epileptic discharge labels; and training the convolutional neural network via the sample electroencephalogram pictures, so that the trained convolutional neural network is obtained. The recognition method and the recognition system provided by the invention have the advantages of being simple in process, low in computation amount, high in recognition rate and free from influence of environmental conditions.

Description

technical field [0001] The invention belongs to the field of pattern recognition of images, and more specifically relates to a convolutional neural network-based epilepsy EEG recognition method and system. Background technique [0002] With the continuous development of medical technology and the continuous improvement of intelligent algorithms, medical technology has a greater demand for more intelligent technology, and intelligent medical systems have become a hot issue in social life. [0003] Most of the existing epilepsy recognition methods use feature extraction and wavelet transform methods. Since epilepsy has various waveforms and is easily disturbed by ECG and EMG, such methods are easily affected by the environment and lead to recognition. Some methods for epileptic EEG recognition have been proposed, especially at present, the mother wavelet template similar to the epileptic discharge waveform is widely used for wavelet transformation, so as to highlight the spike...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the above defects or improvement needs of the prior art, the present invention provides a convolutional neural network-based epilepsy EEG recognition method and system, thereby solving the existing problems of the prior art with complex process, large amount of calculation, and low recognition rate , technical issues that are susceptible to environmental conditions

Images