Classification of EEG signals using genetic programming for feature construction

Abstract

The analysis of electroencephalogram (EEG) waves is of critical importance for the diagnosis of sleep disorders, such as sleep apnea and insomnia, besides that, seizures, epilepsy, head injuries, dizziness, headaches and brain tumors. In this context, one important task is the identification of visible structures in the EEG signal, such as sleep spindles and K-complexes. The identification of these structures is usually performed by visual inspection from human experts, a process that can be error prone and susceptible to biases. Therefore there is interest in developing technologies for the automated analysis of EEG. In this paper, we propose a new Genetic Programming (GP) framework for feature construction and dimensionality reduction from EEG signals. We use these features to automatically identify spindles and K-complexes on data from the DREAMS project. Using 5 different classifiers, the set of attributes produced by GP obtained better AUC scores than those obtained from PCA or the full set of attributes. Also, the results obtained from the proposed framework obtained a better balance of Specificity and Recall than other models recently proposed in the literature. Analysis of the features most used by GP also suggested improvements for data acquisition protocols in future EEG examinations.

A Review of Genetic Programming: Popular Techniques, Fundamental Aspects, Software Tools and Applications

“…Researchers from image processing field have used the GP methods in image processing studies such as in noise suppression [111][112], image reconstruction [113], feature extraction [114], image classification [115] etc. Signal processing: GP algorithms has been utilized in the classification of EEG signals [116], which is a very important task in the diagnosis of several diseases and disorders such as epileptic seizures [117], sleep disorders [141]. One may also see GP employments in processing of other medical signals such as classification of electrocardiography (ECG) signals [20], [118], which are the medical signals that are used to diagnose heart problems.…”

Region-Wise Brain Response Classification of ASD Children Using EEG and BiLSTM RNN

“…17,18 EEG is widely useful in other neurodevelopmental disorders as well such as classification of sleep stages, epileptic identification, emotion classification, seizure detection, early detection of Alzheimer's disease, sleep spindles, and lie detector. [19][20][21][22][23][24][25][26][27] EEG signals are nonlinear, noisy, and have temporal and spatial covariance hence, the right features are to be recognized to attain loss-free information. 26,28 The basic criterion for nonlinear analysis is to represent trajectories in phase space.…”

“…[19][20][21][22][23][24][25][26][27] EEG signals are nonlinear, noisy, and have temporal and spatial covariance hence, the right features are to be recognized to attain loss-free information. 26,28 The basic criterion for nonlinear analysis is to represent trajectories in phase space. Recurrences of system trajectories in the phase space are the best method for systems with nonlinear and nonstationary behavior.…”

“…19–27 EEG signals are nonlinear, noisy, and have temporal and spatial covariance hence, the right features are to be recognized to attain loss-free information. 26,28…”

Evolving boolean functions with conjunctions and disjunctions via genetic programming

“…Despite the many examples of successful applications of GP (see e.g., Koza, 2010;Liu and Shao, 2013;Bartoli et al, 2014;Moore et al, 2018;Miranda et al, 2019;Lynch et al, 2019;Vu et al, 2019), our understanding of its behaviour and performance is limited.…”