Siti Armiza Mohd Aris

EEG-based Stress Features Using Spectral Centroids Technique and k-Nearest Neighbor Classifier

This paper presents the combination of electroencephalogram (EEG) power spectrum ratio and Spectral Centroids techniques to extract unique features for human stress from EEG signals. The combination of these techniques was able to improve the k-NN (k-Nearest Neighbor) clasifier accuracy to detect and classify human stress from two cognitive states, Close-eye (CE) and Open-eye (OE). The EEG power spectrum in term of Energy Spectral Density (ESD) for each frequency bands (Delta, Theta, Alpha and Beta) was calculated. The ratio of EEG power spectrum and the average value of Spectral Centroids were selected as features to k-Nearest Neighbor (k-NN). The training and testing of the classifier were evaluated at 50:50 ratios and 70:30 ratios. The results showed that the combination of EEG power spectrum and Spectral Centroids techniques with the training and testing of k-NN set at 70:30 able to detect and classify the unique features for human stress at 88.89% accuracy.

Evaluation of human stress using EEG Power Spectrum

This paper presents an evaluation conducted between human stress questionnaires with EEG Power Spectrum of Beta and Alpha band. Cohens Perceived Stress Scale (PSS) was used as stress questionnaires to evaluate human stress. The EEG recording of 13 volunteers were carried out immediately after them answering the stress questionnaires. The scores from the stress questionnaires were calculated and used to figure out its relationship with the ratio of EEG Beta and Alpha band power. The results of the study showed the PSS was negatively correlated with the ratio of EEG Power Spectrum. Besides, the study suggested that it was feasible to use PSS and the ratio of EEG Power Spectrum to determine human stress.

Feature Extraction of EEG Signals and Classification Using FCM

EEG data were collected between two conditions, relax wakefulness (close-eyes) and non-relax (IQ test). Data segmentation and linear regression model is used to extract the EEG features and to obtain the slope and the mean relative power from 43 participants. All of the data were then normalized and classified using Fuzzy C-Means (FCM) clustering. Results shown that there are different of activities exist in the EEG which proved that the feature extraction using linear regression model manage to discern between two different brain behaviors.

Stress features identification from EEG signals using EEG Asymmetry & Spectral Centroids techniques

This paper presents EEG Asymmetry and Spectral Centroids techniques in extracting unique features for human stress. The study involved 51 subjects (27 males and 24 females) for Close-eye state (do nothing) and 50 subjects (21 males and 29 females) for Open-eye state (perform IQ test). The subjects then were categorized into 2 groups for all EEG frequency bands (Delta, Theta, Alpha and Beta) by using EEG Asymmetry technique. The negative asymmetry was labelled as Stress group and positive asymmetry was labelled as Non-Stress group. The data in each group in term of Energy Spectral Density (ESD) were normalized by using Z-score technique to produce an index to each asymmetry group. Next, the Spectral Centroids techniques were applied to each group and EEG frequency bands to obtain Centroids values. Since there were 2 asymmetry groups per EEG frequency bands, a total of 8 Centroids values were produced for each cognitive states. The plot of Centroids for both cognitive states showed some unique patterns related to stress.

The Analysis of EEG Spectrogram Image for Brainwave Balancing Application Using ANN

The purpose of this paper is to analysis EEG spectrogram image using Artificial Neural Network (ANN) for brainwave balancing application. Time-frequency approach or spectrogram image processing technique is used to analyze EEG signals. The Gray Level Co-occurrence Matrix (GLCM) texture feature was extracted from spectrogram image and passed through Principal components analysis (PCA) to reduce the feature dimension. The experimental result shows that ANN was able to analysis EEG spectrogram images with an optimized model in training by varying neurons in the hidden layer, learning rate and momentum.

The preliminary study on the effect of nasyid music and rock music on brainwave signal using EEG

This preliminary study analyzes the effect of nasyid music and rock music on brainwave signal particularly focusing on alpha wave. EEG data were recorded from 30 students from Faculty of Electrical Engineering, UiTM age 18 to 27 years old. Students were interviewed before EEG recording to find out their music preference. Using EEG, the brainwave signal of the sample is captured twice, once before listening to the music and while listening to the music. Consequently, the brainwaves signal is analyzed and the comparison between these two music genres is discussed. The results demonstrate that 60–80% of the samples show improvement in the alpha band after listening to nasyid while only 56–66% improves after listening to rock music. These findings indicate that the alpha power increases when listening to the nasyid music compare to the rock music. Consequently, nasyid music in particular can result in a more relaxing condition compared to the rock music

Initial investigation on alpha asymmetry during listening to therapy music

Several studies had shown that frontal alpha-asymmetry is closely related to emotions and motivation. The present study is to investigate the frontal alpha asymmetry during listening to the therapy musical piece. Electroencephalograph (EEG) was used to investigate the emotions states of relax, tense, sad or not focus. 10 participants were involved in this study. Mean power of both hemispheres was computed and the Asymmetry Relation Ratio (ARR) was calculated. From the calculation, individuals with relatively positive ratio shows more relax than the negative ratio.

Performance measure of the multi-class classification for the EEG calmness categorization study

This study presents a small part of the major study, involved in categorizing EEG calmness. The kNN classifier was used to classify EEG features named as asymmetry index (AsI) which was extracted during relaxed state and non-relaxed state. Results from the previous study showed that the EEG behaviour during both states appear to have more than two groups. The group of four EEG behaviours and three EEG behaviours which was clustered by FCM was validated through kNN. However, to investigate the kNN classification accuracy, the classifier performance measure is essential. Thus for this study purposes, performance measure of the kNN was tested using confusion matrix. Result of performance measure indicates that kNN provide 100% accuracy on three clusters of behaviours which could be proposed as calmness index.

Intelligent System for Assessing Human Stress Using EEG Signals and Psychoanalysis Tests

This paper presents a results of designing an intelligent system to evaluate human stress level using Electroencephalogram (EEG) signals and Psychoanalysis tests. The questionnaires for Psychoanalysis tests were created based on Cohens Perceived Stress Scale (PSS). EEG signals were captured using wireless EEG equipment. The Graphical User Interface (GUI) for the Psychoanalysis tests and EEG signals were created. The system was evaluated for 12 healthy subjects (7 females and 5 males). The results show that the intelligent system able to display the stress score, stress level and dominant index of EEG signals simultaneously. Thus, users can use the results of the system to take necessary action in order to improve their lifestyle.

The relationship of alpha waves and theta waves in EEG during relaxation and IQ test

Study on the affiliation of the EEG alpha waves and theta waves during relaxation state and IQ test state from sixteen college students has been carried out in this paper. The alpha waves and theta waves are very much related to the relaxation and concentration respectively. By using data segmentation and linear regression techniques, power changes in alpha band and theta band are plotted. It is revealed that the alpha band relative power values reduced when the IQ test session is held and it is found that the Index 6 from the IQ group produced highest response of alpha and theta mean power comparing to other IQ Indexes. It is shown that the relative power segmentation techniques in conjunction with linear regression are able to shows significant dissimilarity among the IQ Index group by relating relaxation and IQ test experiments.