Spectral entropy feature subset selection using SEPCOR to detect alcoholic impact on gamma sub band visual event related potentials of multichannel electroencephalograms (EEG)
Graphical abstract
Introduction
It is well-established that there will be structural and functional changes in the neuronal activities of human brain upon consumption of alcohol and even after long term abstinence [1]. Event related potentials (ERP) exhibit cerebral activity that characterize spatio-temporal changes in the human brain over a period of time due to alcoholism [2], [3], [4]. The dynamic processes of brain such as memory, attention and cognitive processing [5], [6] are correlated with synchronizations of phase locked peaks generated by ERP which are characterized as delta, theta, alpha, beta and gamma waves. Changes in characteristics of these waves due to alcoholism are reported extensively in literature [7], [8], [9], [10], [11].
Numerous studies have reported the use of linear, non-linear and statistical measures (features) for the classification of alcoholic and control EEG epochs achieving very high classification accuracies. The extraction of power spectral density (PSD) of the γ band (> 30 Hz) has been extensively used for classification of alcoholics and controls in the literature using FFT and other parametric techniques [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23]. The γ sub band between 30–50 Hz is assumed to contain the visually evoked responses that differentiate an alcoholic from a control subject. These studies confirmed that the parametric methods were superior in their performance as compared to FFT due to their better frequency resolution, leading to achieve better SNR accuracy. Also among all the parametric methods used, Burg’s method was reported to extract the most distinguishable parameters for classifying alcoholics and controls
Studies have also shown that there is a reduction in evoked γ oscillations [3] in alcoholics during the processing of a visual object recognition task in the occipital and frontal regions. The deficits in γ oscillations manifest themselves as cognition deficits in selective attention and working memory tasks. In order to study cognitive impairments in alcoholics, it is necessary to identify specific regions of brain that are largely influenced by alcohol consumption during these event related oscillations. Therefore selection of channels with high discrimination between alcoholic and control subjects using prominent features extracted from visual ERP’s of a multichannel EEG recording in γ band needs to be investigated. Selection of optimal subset from 64 channel EEG recordings of alcoholics and controls has been explored by a few researchers [15], [19], [24], [25], [26], [27], [28], [29]. The spectral power ratios in multiple sub gamma bands and mean gamma band power in the visual evoked potentials of multichannel EEG signal are reported to be good discriminators for optimal feature selection and classification of alcoholics and controls [14], [25]. An interesting study [15], reported on the selection of seven out of 61 channels using genetic algorithm (GA) optimization to effectively discriminate alcoholics and controls. Within the entire spectrum of 2–50 Hz, only γ2 (37–43 Hz) and γ3 (44–50 Hz) bands of the seven optimized channels were reported to be effective features providing an average classification accuracy of 94.3% and 81.8% with MLP-BP network and fuzzy art map (FAM) classifier respectively. Studies by Ong et al. [24] have shown that the use of PCA (Principal Component Analysis) for selecting optimal channels to detect alcoholics resulted in classification accuracies of 95.83%, 94.06%, 86.01% and 75.13% for 61, 16, 8, and 4 channels respectively. The disadvantage here is that the use of PCA requires large memory while handling large datasets.
The correlation between selected optimal subset of channels for alcoholics was not explored until a specific study [25] that was carried out to select a set of active channels based on mean γ band power for each channel per subject. These active channels were later used to determine correlation coefficient among them and only those which scored a correlation coefficient greater than 0.6 with other selected channels were retained as candidate channels to train least square SVM classifier for the discrimination of alcoholics and controls. An average classification accuracy of 80% was reported among different pairs of 45 optimal channels. The problem in this method is that the highly correlated channels become redundant as far as classification accuracy is considered which may also lead to over fitting.
A study [19] using non-linear parameters such as ApEn (Approximate Entropy), Sample Entropy, Lyapunov Exponent and Higher order spectra for feature extraction and detection of alcoholics reported an accuracy of 91.7% using SVM classifier. The selection of channels was done by performing a statistical t-test on all the features and seven of them are used for classification The study used the same alcohol EEG dataset [30] as in the proposed study with 30 alcoholic and 30 control subjects visual ERP’s.
In all these studies, the class information between groups is not considered while choosing features. Even though the selection of optimal channel subset is based on certain criterion such as the gamma band power or the statistical t-test or correlation among them, the merit of each individual channel within the subset is not weighed in terms of its ability to separate alcoholics from controls. Hence the main objective of this proposed research is to choose only those optimal spectral entropy features which maximize the separation between classes with low correlation among them by using a statistical pattern recognition method called Separability and Correlation (SEPCOR) method [31], [32], [33]. It computes a variability measure (V-measure) for each channel which is defined as a ratio of the variance of class means to the mean of variance within the class. A high V-measure for a channel implies that the variance between class outputs is large while the mean of variance within the class output is small. Thus the high V-measure of a channel is indicative of maximum class separability and minimum correlation value and hence more discriminative of groups. The channels are arranged in a descending order with respect to their V- measure. A feature with highest V-measure is the first candidate feature chosen and the second feature is selected if its correlation coefficient with the first chosen channel is less than a preset (user defined) correlation threshold called, MAXCOR else it will be rejected. Proceeding further, the next sorted channel is tested for correlation threshold with the previously selected channel with highest V-measure. This method is repeated until all the features are exhausted. The channels selected thus represent a minimum set of (subset) highly uncorrelated channels with large V-measures The validity of the proposed method for feature selection is evaluated by applying SEPCOR reduced feature set as inputs to MLP and k-NN classifiers. The classifier performances are cross validated using 50% hold out cross validation.
Recently, entropy estimation of EEG signals have been used extensively to explain how EEG signals change with time either in frequency or in phase domain [34], [35], [36]. It implies many possible different rates of change of summed pyramidal cell membrane potentials. Commercial entropy monitors measure the depth of anesthesia as EEG activity is more regular (orderly) under its influence than in wakefulness [37], [38]. Higher values of entropy are indicative of more complexity involved in the dynamic processes of different cortical regions. It is reported that the spectral peak of increasing amplitude in α-frequency band causes spectral entropy to increase when low frequencies are present in the signal [39]. Thus the entropy estimation provides a measure of disorderliness and hence some important information regarding the complexity of processes involved in a system. These studies lead us to explore the potential use of entropy as an effective discriminating feature for classifying alcoholic/control visual ERP in sub gamma band range. In the present study, Spectral Entropies (SE) of artifact free EEG recordings are estimated in the gamma sub band range (30–55 Hz) for both alcoholic and control subjects’ 61-channel EEG recordings. The estimated SE features from 61 channels of both alcoholic and control gamma sub band visual ERP form the entire feature set. Traditionally, EEG data epochs containing any or/and all of these artifacts are rejected and this may lead to loss of important information from the dataset. To overcome this, artifacts are separated from actual EEG signals by performing linear transformations using methods such as Principal Component Analysis (PCA), Independent Component Analysis (ICA) etc. In the current study, ICA is used to separate EEG signal from eye blink, motion and muscle artifacts.
The rest of the paper is organized as follows: Section 2 describes the proposed methodology. Results and discussion are reported in Section 3 and 4 respectively. Conclusion is reported in Section 5.
Section snippets
The proposed method
In our earlier work [40], despite obtaining good classification accuracy, significant channels were not identified which contribute to the discrimination of alcoholics from their control counterparts. Hence in the proposed work, selection of an optimal feature subset is performed by using SEPCOR algorithm to study the impact of alcohol on specific regions of the brain that effectively discriminate the visual ERP of alcoholics from controls. Fig. 1 shows schematic diagram of the steps involved
Results
Initially during the training session of MLP network and k-NN classifier, out of 600 alcoholic and 600 control spectral entropy vectors, 60% is used for training using Matlab custom function. Each training cycle out of the 600 training epochs randomly chooses 60% training vectors to achieve better generalization. During testing session with untrained data as input, 50% holdout cross validation is performed for 10 runs. Each run randomly chooses 50% of the alcoholic and control epochs for cross
Discussion
The SEPCOR selected channels enhance the performance of the classifier drastically thus proving that the use of class information for optimum feature selection prior to classification indeed improves the classification accuracy. The classification accuracy vastly improves in the case of MLP network even with 5 selected channels as shown in Fig. 12. The user defined correlation threshold, MAXCOR, automatically chooses the number of channels with maximum class separation and minimum correlation
Conclusion
Compared to feature selection methods used in previous studies using the same alcoholic/control EEG database, the accuracy of detection improves considerably using the proposed SEPCOR method. The proposed feature selection method not only rejects features with maximum correlation but also selects those features which maximize class separation. Due to this, the classifier is provided with a better feature representation at the input. It is also observed that the channel 12 corresponding to FC2
Acknowledgements
We thank Prof. Henri Begleiter at the Neuro dynamics Laboratory at the State University of New York Health Center at Brooklyn, USA for sharing the EEG database in the public domain. We are also thankful to Dr. Vivek Benegal, Dept. of Psychiatry, NIMHANS, Bangalore for his invaluable guidance. The authors would like to thank the anonymous reviewers for their helpful comments and suggestions that greatly improved the quality and clarity of the manuscript.
References (62)
- et al.
Chronic alcoholism: insights from neurophysiology
J. Clin. Neurophysiol.
(2009) - et al.
A review of brain oscillations in cognitive disorders and the role of neurotransmitters
Brain Res.
(2008) - et al.
Delta and theta oscillations as risk markers in adolescent offspring of alcoholics
Intl. J. Psychophysiol.
(2007) - et al.
Induced theta oscillations as biomarkers for alcoholism
J. Clin. Neurophysiol.
(2010) - et al.
Electrophysiological evidence of memory impairment in alcoholic patients
J. Biol. Psychiatry
(1997) - et al.
Using genetic algorithm to identify the discriminatory subset of multi-channel spectral bands for visual response
J. Appl. Soft Comput.
(2002) - et al.
Analysis of EEG signals during epileptic and alcoholic states using AR modeling techniques
J. Innov. Technol. Biol. Med.
(2008) - et al.
Characterization of EEG—a comparative study
Comput. Methods Programs Biomed.
(2005) - et al.
Alcoholism-related alteration in spectrum, coherence, and phase synchrony of topical electroencephalogram
Comput. Biol. Med.
(2012) - et al.
Event related potentials during object recognition tasks
Brain Res. Bull.
(1995)
Wavelet entropy analysis of event-related potentials indicates modality-independent theta dominance
J. Neurosci. Methods
Recovering EEG brain signals: artifact suppression with wavelet enhanced independent component analysis
J. Neurosci. Methods
Removal of eye activity artifacts from visual event-related potentials in normal and clinical subjects
J. Clin. Neurophysiol.
Coarse to fine K nearest neighbor classifier
Pattern Recognit. Lett.
Hypo perfusion of inferior frontal brain regions in abstinent alcoholics: a pilot SPECT study
J. Stud. Alcohol
Alcoholism and human electrophysiology
Alcohol Res. Health
Positron emission tomography—a tool for identifying the effects of alcohol dependence on the brain
Alcohol Res. Health
Attention deficits in alcoholic brain syndrome
Annual IEEE Intl. Conf. on Engg in Medicine and Biology Society
Association of low- voltage alpha EEG with a subtype of alcohol use disorders
J. Alcohol Clin. Exp. Res.
Screening for chronic alcoholic subjects using multiple gamma band EEG: a pilot study
Int. J. Comput. Sci. Technol.
Discrimination of alcoholic subjects using second order autoregressive modeling of brain signals evoked during visual stimulus perception
J. World Acad. Sci. Eng. Technol.
Improved automated classification of alcoholics and non-alcoholics
J. Inf. Technol.
Screening for chronic alcoholic subjects using multiple gamma band EEG: a pilot study
J. Comput. Sci. Technol.
Auto associative neural networks for discrimination of chronic alcoholics using visual evoked potentials
Intl. Conf. on Computing Communication and Networking
Fuzzy ART map classification of alcoholics and non-alcoholics based on visual evoked potential
Intl. Conf. on AEESEAP
Automated diagnosis of normal and alcoholic EEG signals
J. Neural Syst.
Low doses of ethanol reduce evidence for nonlinear structure in brain activity
J Neurosci
Computer-based identification of normal and alcoholic EEG signals using wavelet packets and energy measures
J. Mech. Med. Biol.
Selection of a subset of EEG channels using PCA to classify alcoholics and non-alcoholics
Proc. of IEEE Conf. on EMBS
Selection of optimal EEG channels for classification of signals correlated with alcohol abusers
Proc. of the IEEE 10th Intl. Conf. on Signal Processing
Neural, fuzzy and neuro fuzzy approach to classification of normal and alcoholic electroencephalograms
Proc. of 5th Intl. Symposium on Image and Signal Processing and Analysis
Cited by (27)
EEG-based classification of individuals with neuropsychiatric disorders using deep neural networks: A systematic review of current status and future directions
2023, Computer Methods and Programs in BiomedicineClassification of focal and non-focal EEG signals using neighborhood component analysis and machine learning algorithms
2018, Expert Systems with ApplicationsCitation Excerpt :The advantage of NCA is that it provides the information of significant features along with feature ranking. Non-linear classifiers such as SVM and K-NN models have been widely used for EEG pattern classification problems due to the learning ability of the supervised model to discriminate binary classes (Bogaarts, 2016; Jaiswal & Banka, 2017; Padma Shri & Sriraam, 2016a; 2016b; 2017; Sharma et al., 2017; Sharma et al., 2015a; 2015b; Sriraam & Raghu, 2017; Subasi & Gursoy, 2010). Although random forest and Adaboost classifiers have applied for seizure classification, attempts have not been made for focal and non-focal EEGs classification.
Comparison of t-test ranking with PCA and SEPCOR feature selection for wake and stage 1 sleep pattern recognition in multichannel electroencephalograms
2017, Biomedical Signal Processing and ControlCitation Excerpt :In feature subset selection methods like PCA, the features are in a transformed space whereas in SEPCOR, features remain in their original space. The numerator in Eq. (8) measures the distance between class means whereas the denominator represents the spread of class means. [48,54]. A high value of V-measure is desirable for a feature to be considered as a suitable candidate for classifier input because it indicates that the variance between class means is large and the mean of variance within class is small.
Deep clustering using 3D attention convolutional autoencoder for hyperspectral image analysis
2024, Scientific ReportsGrey Level Differences Matrix for Alcoholic EEG Signal Classification
2024, International Journal on Informatics Visualization