Adaptive multi-parent crossover GA for feature optimization in epileptic seizure identification
Introduction
There are various types of neurological diseases such as neurogenetic diseases, degenerative diseases and convulsive disorders [1]. Most of the convulsive disorders occur as a result of an uneven electrical conductivity in the brain which results to uncontrollable quivering of the body. Despite the fact that seizures are part of epilepsy, not all seizures are as a result of epilepsy [2]. Epilepsy affects about 1% of the world population [3]. It is one of the most common neurological diseases, which is incurable, occurs recurrently and unpredictable [4], [5]. However, epilepsy can be controlled with medications, and surgery when patient does not respond to medication [6]. Electroencephalography (EEG) is a brain signal processing technique that provides an insight to the complex inner mechanism of normal and abnormal brain waves, which is used to diagnose brain disorder [7]. Practically, epileptic seizures can be seen as an abnormal, automatic movement or alteration of consciousness related to abnormal EEG changes [8]. However, the EEG signal characteristics differ among patients and between different seizures in the same patient. Since epilepsy show abnormal EEG signal changes, clinically intra-cranial EEGs are used to diagnose, differentiate and classify epileptic seizures [9]. These signals can appear in different forms such as spikes, poly-spikes, or spike and waves. EEG recordings are gathered during seizures and between seizures. These huge data are currently interpreted manually by medical personals, making the process of visual analysis cumbersome. The main aim of a computer-aided EEG classification is not only to reduce the time and effort for medical personals but also to be able to predict future occurrences of epileptic seizures.
Relevant information regarding different brain disorder and states are provided by the EEG signals. These signals are collected in form of datasets and transferred to tools for plotting and visualization of the time–frequency plots [7]. The challenges of automating this process for seizure detection includes:
- 1.
The variety of signal characteristics that vary between patients (inter-patient variability), along with different seizures in the same patient (intra-patient variability).
- 2.
Abnormalities seen during epileptic seizures can also be observed during some non-epileptic seizures.
- 3.
Addressing the high complexity and nonstationarity heterogeneous types seizures [10].
- 4.
The EEG includes massive amounts of data and handling them is a costly process [11].
The aim of this work is to propose a technique for epileptic seizure identification in order to efficiently analyze EEG signals for identifying epileptic and non-epileptic seizures. This complex problem of classifying aberrant EEG signals warrants the use of improved feature extraction and reduction techniques, and classification algorithms. The proposed technique adopts a GA, with an adaptive multi-parent crossover, to encode the temporal and spatial filter estimates and optimize the feature selection of EEG.
Genetic algorithms, as a member of the evolutionary algorithms, are search algorithms that imitates the natural evolution process. It defines how a population can regenerate a modified version of its individuals in order to satisfy the required goal based on the natural concept; survival of the fittest. The GA works on coded population by an initializing process. Then a selection process is applied to select some solution that are evaluated to find their fitness value. After that, a reproduction process starts to generate new offspring using crossover and mutation operators. Finally, the replacement process replaces the old solutions by new ones. For more information the reader may refer to [12].
The rest of the paper is organized as follows: In Section 2, a brief survey and a review of the different implementations of epileptic seizure classification is presented. Section 3 presents the proposed framework that comprises of four phases; namely, the windowing phase, feature extraction phase, optimization phase, and classification. The materials and experimental methods are highlighted in Section 4. In Section 5, the results are presented and discussed in Section 6. Finally, Section 7 concludes the research with the findings and future work.
Section snippets
Related works
The major challenge of automated detection of epileptic seizures involves the use of techniques that can classify these seizures in consideration of both inter-patient and intra-patient variability. The basic approach used in identifying epileptic seizure involves feature extraction, feature normalization and seizure classification. The EEG signals are first decomposed into sub-bands. Afterwards, relevant features are extracted and normalized using several approaches before a classifier is used
The proposed approach
EEG signal analysis involves multi-frequency non-stationary brain waves from multiple channels. Segmenting these signals, extracting features to obtain the important properties of the signal and classification are important aspects of detecting epileptic seizures. Despite the introduction of several techniques, it is very challenging when multiple EEG channels are involved. When many channels exist, a spatial filter is required to eliminate noise and extract relevant information. In order to
Materials and method
In this section, we outline the experimental method, evaluation criteria and the data used for evaluating the proposed framework. The evaluation of our proposed method for the epileptic seizure identification is done on the 2017 re-structured dataset provided by the machine learning repository of the UCI center for machine learning and intelligent systems [40]. The dataset consists of five different folders, with 100 files each representing a single subject. Thus, resulting to 500 subjects.
Results
The main purpose of feature selection is to reduce the feature vector space and still maintain a good classification accuracy. A reduced feature ensures a faster training and classification time. The proposed multi-parent crossover GA approach provides a reduced feature vector space and enhanced accuracy. As shown in Fig. 3, the convergence of the fitness function towards the maximum accuracy is shown through 100 generation. During this optimization phase an accuracy of 98.98% was obtained. The
Discussion
The results obtained from the evaluation indicates a positive effect of the MPC-GA technique on the optimization of the feature selection process. A better accuracy and fewer features are found using our approach. Based on the feature ranking in Table 5, random features selected from 89–92 showed considerably higher accuracy than order randomly chosen features. This indicates the effect of the AMPC GA technique is selecting optimal number of features for classification. Some of the previous
Conclusion
The uncontrollable and erratic nature of epileptic seizures is a concern in medicine. Currently, the process of diagnosis epileptic seizure is complicated and many mis-diagnosis occur. However, the EEG technique provides an insight into a patients brain signals. Epileptic seizures can be seen as abnormal changes in the EEG. But the EEG signal abnormality defers among patients. The automated process of accurately analyzing this data from the EEG signals require a huge dataset in order to improve
References (40)
- et al.
Automatic epileptic seizure detection in EEGs based on line length feature and artificial neural networks
J. Neurosci. Methods
(2010) EEG signal classification using wavelet feature extraction and a mixture of expert model
Expert Syst. Appl.
(2007)- et al.
EEG signals classification using the K-means clustering and a multilayer perceptron neural network model
Expert Syst. Appl.
(2011) - et al.
Detection of epileptic electroencephalogram based on permutation entropy and support vector machines
Expert Syst. Appl.
(2012) - et al.
Hilbert marginal spectrum analysis for automatic seizure detection in EEG signals
Biomed. Signal Process. Control
(2015) - et al.
MMSFL-OWFB: A novel class of orthogonal wavelet filters for epileptic seizure detection
Knowl.-Based Syst.
(2018) - et al.
Viral vector gene delivery to the brain for treating neurogenetic diseases
- et al.
Epilepsy, psychiatry, and neurology
Epilepsia
(2009) - et al.
Time domain analysis of EEG signals for detection of epileptic seizure
- et al.
Antiepileptic drug use in Italian children over a decade
Eur. J. Clin. Pharmacol.
(2017)
Canine versus human epilepsy: Are we up to date?
J. Small Anim. Pract.
Refractory epilepsy in children with brain tumors. The urgency of neurosurgery
Arq. Neuro-psiquiatr.
Analysis and simulation of brain signal data by EEG signal processing technique using MATLAB
Int. J. Eng. Technol. (IJET)
Introduction to Epilepsy
Automated classification of EEG signals using component analysis and support vector machines
Performance reassessment of a real-time seizure-detection algorithm on long ECoG series
Epilepsia
Genetic algorithms
Hierarchical multi-class SVM with ELM kernel for epileptic EEG signal classification
Med. Biol. Eng. Comput.
Detection of seizure and epilepsy using higher order statistics in the EMD domain
IEEE J. Biomed. Health Informatic.
Multi-task seizure detection: Addressing intra-patient variation in seizure morphologies
Mach. Learn.
Cited by (28)
A Genetic Algorithm Assisted Fuzzy Q-Learning epileptic seizure classifier
2021, Computers and Electrical EngineeringCitation Excerpt :Sensitivity, Specificity and Accuracy achieved is 91.9%, 93.2% and 94.0%, respectively. Al-Sharhana et al. [22] showcased a multi-parent crossover GA for optimization of epileptic seizure classification. It uses spatial and temporal estimates for improving classification performance.
An energy screening and morphology characterization-based hybrid expert scheme for automatic identification of micro-sleep event K-complex
2021, Computer Methods and Programs in BiomedicineCitation Excerpt :Inspired by the Chambon et al. method, one way to improve our approach in future work would be to automate parameter setting for data through optimization techniques such as the genetic algorithm. Like the neural network, the genetic algorithm is one of the heuristic methods and has been applied successfully in biomedical signal processing [42]. By introducing the optimization algorithm into our scheme, the system could be tuned automatically with optimal parameters while retaining system understandability and acceptance by physicians.
A Crossover Optimization Method based on Correlation Coefficient in Differential Evolution
2023, 7th International Conference on Trends in Electronics and Informatics, ICOEI 2023 - ProceedingsResistance coefficient identification of a heating pipe network based on a heuristic three-parent genetic algorithm
2023, Engineering Optimization