In order to improve the accuracy and efficiency of automatic seizure detection, the paper proposes a method based on improved genetic algorithm optimization back propagation (IGA-BP) neural network for epilepsy diagnosis, and uses the method to achieve detection of clinical epilepsy rapidly and effectively. Firstly, the method extracted the linear and nonlinear features of the epileptic electroencephalogram (EEG) signals and used a Gaussian mixture model (GMM) to perform cluster analysis on EEG features. Next, expectation maximization (EM) algorithm was used to estimate GMM parameters to calculate the optimal parameters for the selection operator of genetic algorithm (GA). The initial weights and thresholds of the BP neural network were obtained through using the improved genetic algorithm. Finally, the optimized BP neural network is used for the classification of the epileptic EEG signals to detect the epileptic seizure automatically. Compared with the traditional genetic algorithm optimization back propagation (GA-BP), the IGA-BP neural network can improve the population convergence rate and reduce the classification error. In the process of automatic detection of epilepsy, the method improves the detection accuracy in the automatic detection of epilepsy disorders and reduced inspection time. It has important application value in the clinical diagnosis and treatment of epilepsy.
It is very important for epilepsy treatment to distinguish epileptic seizure and non-seizure. In this study, an automatic seizure detection algorithm based on dual density dual tree complex wavelet transform (DD-DT CWT) for intracranial electroencephalogram (iEEG) was proposed. The experimental data were collected from 15 719 competition data set up by the National Institutes of Health (NINDS) in Kaggle. The processed database consisted of 55 023 seizure epochs and 501 990 non-seizure epochs. Each epoch was 1 second long and contained 174 sampling points. Firstly, the signal was resampled. Then, DD-DT CWT was used for EEG signal processing. Four kinds of features include wavelet entropy, variance, energy and mean value were extracted from the signal. Finally, these features were sent to least squares-support vector machine (LS-SVM) for learning and classification. The appropriate decomposition level was selected by comparing the experimental results under different wavelet decomposition levels. The experimental results showed that the features selected in this paper were different between seizure and non-seizure. Among the eight patients, the average accuracy of three-level decomposition classification was 91.98%, the sensitivity was 90.15%, and the specificity was 93.81%. The work of this paper shows that our algorithm has excellent performance in the two classification of EEG signals of epileptic patients, and can detect the seizure period automatically and efficiently.