An Epilepsy Detection Method Using Multiview Clustering Algorithm and Deep Features

Abstract

The automatic detection of epilepsy is essentially the classification of EEG signals of seizures and nonseizures, and its purpose is to distinguish the different characteristics of seizure brain electrical signals and normal brain electrical signals. In order to improve the effect of automatic detection, this study proposes a new classification method based on unsupervised multiview clustering results. In addition, considering the high-dimensional characteristics of the original data samples, a deep convolutional neural network (DCNN) is introduced to extract the sample features to obtain deep features. The deep feature reduces the sample dimension and increases the sample separability. The main steps of our proposed novel EEG detection method contain the following three steps: first, a multiview FCM clustering algorithm is introduced, and the training samples are used to train the center and weight of each view. Then, the class center and weight of each view obtained by training are used to calculate the view-weighted membership value of the new prediction sample. Finally, the classification label of the new prediction sample is obtained. Experimental results show that the proposed method can effectively detect seizures.

Predicting cognitive impairment in outpatients with epilepsy using machine learning techniques

“…Until now, most of the exploration in epilepsy with cognitive impairment has used traditional advanced statistical methods, such as LR 29 . These methods have an overt restriction on the sample size of the research population.…”

Epileptic seizure prediction based on Convolutional neural networks and optimization techniques

“…Accuracy Specificity Sensitivity EDMLC [42] 94.12 94.87 94.02 FCM-MPSO [43] 93.17 92.05 92.01 DCNN-FCM [41] 97 Future seizure prediction techniques should be developed with the intention of being patient-independent and effective for epilepsy patients.…”

Machine Learning Algorithms for Epilepsy Detection Based on Published EEG Databases: A Systematic Review

“…According to the proposed methodology, the Freiburg DB EEG recordings are cut into 4-second periods and a 5-level DWT is applied, the high frequency coefficients (>32 Hz) are removed and the signal is reconstructed. Zhan et al [123] utilized DWT along with Fourier Transform and a Convolution Block for feature extraction and trained a Fuzzy C-means classifier achieving ACC = 89.75%, SENS = 85.52% in the ictal-interictal problem. Mu et al [124] combined DWT with a Graph-regularized non-negative matrix factorization for 21 patients with intracable focal epilepsy and trained a Bayesian LDA classifier achieving ACC = 98.16%, SENS = 93.2%, SPEC = 98.16% in the same problem.…”