A Novel Use of Discrete Wavelet Transform Features in the Prediction of Epileptic Seizures from EEG Data

Abstract

This paper demonstrates the predictive superiority of discrete wavelet transform (DWT) over previously used methods of feature extraction in the diagnosis of epileptic seizures from EEG data. Classification accuracy, specificity, and sensitivity are used as evaluation metrics. We specifically show the immense potential of 2 combinations (DWT-db4 combined with SVM and DWT-db2 combined with RF) as compared to others when it comes to diagnosing epileptic seizures either in the balanced or the imbalanced dataset. The results also highlight that MFCC performs less than all the DWT used in this study and that, The mean-differences are statistically significant respectively in the imbalanced and balanced dataset. Finally, either in the balanced or the imbalanced dataset, the feature extraction techniques, the models, and the interaction between them have a statistically significant effect on the classification accuracy. ⋆ This document is the results of the research project funded by AIMS CAMEROON with the help of Mastercard Foundation. In this work, we demonstrate the predictive superiority of discrete wavelet transform over previously used methods of feature extraction in the diagnosis epileptic seizures from EEG data.

Epileptic Seizures Detection from EEG Recordings Based on a Hybrid system of Gaussian Mixture Model and Random Forest Classifier

“…Feature extraction techniques reduce the amount of data that must be processed while still properly and thoroughly characterizing the original dataset by selecting and/or combining variables into meaningful features [10]. This study presents one feature extraction technique applied on the EEG dataset, namely Wavelet Transform (WT).…”

“…Therefore, the development of an automated, computer-aided approach to epilepsy diagnosis is critical [9]. As a result, several approaches for detecting epileptic seizures using EEG recordings have been developed, and Machine-learning algorithms were used for this task, including gathering Electroencephalography (EEG) signals, preprocessing, feature extraction from the data, and ultimately classification of epileptic seizures [10]. In recent years, researchers have attempted to discover a more effective solution in the machine-learning field to increase prediction performance [4].…”

“…Figure 1: Four-level EEG signal decomposition.by four major factors: DWT coefficient characteristic, mother wavelet, frequency band, and decomposition level[10].…”