Neurological Disorders

Epileptic seizure prediction from multivariate sequential signals using Multidimensional convolution network

Abstract

Xiaoyan Wei*, Xiaojun Cao, Yi Zhou and Zhang Zhen

Background: The ability to predict coming seizures will improve the quality of life of patients with epilepsy. Analysis of brain electrical activity using multivariate sequential signals can be used to predict seizures.

Methods: Seizure prediction can be regarded as a classification problem between interictal and preictal EEG signals. In this work, hospital multivariate sequential EEG signals were transformed into multidimensional input, multidimensional convolutional neural network models were constructed to predict seizures several channels segments were extracted from the interictal and preictal time duration and fed them to the proposed deep learning models.

Results: The average accuracy of multidimensional deep network model for multi-channel EEG data is about 94%, the average sensitivity is 88.47%, and the average specificity is 89.75%.

Conclusion: This study combines the advantages of multivariate sequential signals and multidimensional convolution network for EEG data analysis to predict epileptic seizures, thereby enabling early warning before epileptic seizures in clinical applications.