Sharif Digital Repository

This research presents a robust method for P300 component recognition and classification in EEG signals for a P300 Speller Brain-Computer Interface (BCI). The multiresolution wavelet decomposition technique was used for feature extraction. The feature selection was done using an improved t-test method. For feature classification the Quadratic Discriminant Analysis was employed. No any particular specification is previously assumed in the proposed algorithm and all the constants of the system are optimized to generate the highest accuracy on a validation set. The method is first verified in offline experiments on 'BCI competition 2003' data set IIb and data recorded by Emotiv Neuroheadset and... 

Migraine is a common type of headache with neurovascular origin. In this paper, a quantitative analysis of spontaneous EEG patterns is used to examine the migraine patients with maximum and minimum pain levels. The analysis is based on alpha band phase synchronization algorithm. The efficiency of extracted features are examined through one-way ANOVA test. we reached the P-value of 0.0001, proving that the EEG patterns are statistically discriminant in maximum and minimum pain levels. We also used a Neural Network based approach in order to classify the EEG patterns, distinguishing between minimum and maximum pain levels. We achieved the total accuracy of 90.9 %  

Modeling is one of assessing tools for better understanding of human body organs and study of diseases. One of the brain diseases is ADHD, which has been studied before, mostly by means of EEG signals. In this paper, the mean-field model, which is a model of neuron-population spiking, and the Power Spectrum of the resulting spikes have been studied by changing parameters of model. The results show that there is a meaningful relationship between firing activity of ADHD patients neuron population and the parameters of mean-field model and Power Spectrum of spikes. In addition, the effects of stimulant medications for ADHD patients on firing activity and power spectrum of firing activity of... 

Epileptic seizures are generated by an abnormal synchronization of neurons. Since epileptic seizures are unforeseeable for the patients, epileptic seizures detection is an interesting issue in epileptology, that novel approaches to understand the mechanism of epileptic seizures. In this study we analyzed invasive electroencephalogram (EEG) recordings in patients suffering from medically intractable focal epilepsy with a nonlinear method called, dissimilarity index. In order to detect epileptic seizures Bhattacharyya distance between trajectory matrix of reference window during an interval quite distant in time from any seizure and trajectory matrix of present window is employed to measure... 

Controlling the environment with EEG signals is known as brain computer interface is the new subject researchers are interested in. The aim in such systems is to control the machine without using muscle, and we should control the machine using signals recorded from the surface of the cortex. In this project our focus is on pattern recognition phase in which we use multiple classifier fusion to improve the classification accuracy. We have applied various feature extraction methods and combined their results. Two methods, greedy algorithms and genetic algorithms, are used for selecting the pair feature extractor-classifier (we called expert) between the existed pair. Experiments show that with... 

This paper presents a scalp electroencephalogram (EEG) rhythmic pattern detection scheme based on neural networks. Rhythmic discharges detection is applicable to the majority of seizures seen in newborns, and is listed as detecting 90% of all the seizures. In this approach some features based on various methods are extracted and compared by a modified multilayer neural network in order to find rhythmic discharges. Statistical performance comparison with seizure detection schemes of Gotman et al. and Liu et al. is performed. © 2006 IEEE  

In this paper, the performance of traditional variance-based method for detection of epileptic seizures in EEG signals are compared with various methods based on nonlinear time series analysis, entropies, logistic regression, discrete wavelet transform and time frequency distributions. We noted that variance-based method in compare to the mentioned methods had the best result (100%) applied on the same database. © 2006 IEEE  

Brain computer interface (BCI) is a widely used system to assist the disabled and paralyzed people by creating a new communication channel. Among the various methods used in BCI area, motor imagery (MI) is the most popular and the most common one due to its the most natural way of communication for the subject. Some software applications are used to implement BCI systems, and some toolboxes exist for EEG signal processing. In recent years virtual reality (VR) technology has entered into the BCI research area to simulate the real world situations and enhance the subject performance. In this work, a completely MATLAB-based MI-based BCI system is proposed and implemented in order to navigate... 

Contamination of ictal Electroencephalogram (EEG) signals by muscle artifacts is one of the critical issues related to clinically diagnosing seizure. Over the past decade, several methods have been proposed in time, frequency and time-frequency domain to accurately isolate ictal EEG activities from artifacts. Among denoising approaches Canonical Correlation Analysis (CCA) and Independent Component Analysis (ICA) are widely used. Denoising based on Generalized EigenValue Decomposition (GEVD) is one of the Semi-Blind Source Separation (SBSS) methods which has been recently proposed. In the GEVD-based method, a couple of time-frequency covariance matrices are used. These time-frequency (TF)... 

Sleep study is very important in the health since sleep disorders affect the productivity of individuals. One of the important topics in sleep research is the classification of sleep stages using the electroencephalogram (EEG) signal. Electrical activities of brain are measured by EEG signal in the laboratory. In real-world environments, EEG signal is also used in portable monitoring devices to analyze sleep. In this study, we propose an efficient method for classification of sleep stages. EEG signals are examined by a new model from autoregressive (AR) family, namely logistic smooth transition autoregressive (LSTAR) to study sleep process. In contrast to the AR model, LSTAR is a non-linear... 

Sleep Spindle is the hallmark of the second stage of sleep in EEG signal. It had been analyzed using different methods, including Fourier transform, parametric and non-parametric models, higher order statistics and spectra, and also time-frequency methods such as wavelet transform, and matching pursuit. In this study, bump modeling has been used to analyze sleep spindle. Bump modeling is a method which represents the time-frequency map of signals with a number of elementary functions. Results of this work demonstrate that bump modeling is capable of analyzing different sleep spindle patterns in sleep EEG signals successfully  

Detection of seizure periods in an epileptic patient is an important part of health care. However, due to the variety in types of seizures and location of them, real-time seizure detection is not straight forward. In this paper, we propose a method for seizure detection from EEG signals in datasets which have both generalized and focal seizures. The proposed method is useful in the situations that we have no prior knowledge about the location of the patient's seizure and the pattern of evolution of seizure location. In the proposed method, first, the artifacts are automatically reduced by Blind Source Separation (BSS) methods. Then, the channels are clustered into two clusters. After that,... 

The efficiency of a learning process directly depends on how well the students are attentive. Detecting the level of attention can help to improve the learning quality. In recent years, there have been several attempts to leverage EEG signal processing as a tool to detect whether a student is attentive or not. In such work, the level of attention is determined by analyzing the EEG power spectrum, which is mostly followed by machine learning approaches. However, the efficiency of such methods for detecting auditory attention of blind or visually-impaired students has not been analyzed. This study aims to investigate such a scenario. To this end, we propose an EEG recording protocol to... 

This study presents a new method for epilepsy detection based on autoregressive (AR) estimation of EEG signals. In this method, optimum order for AR model is determined by Bayesian Information Criterion (BIC) and then AR parameters of EEG signals (from EEG data set of epilepsy center of the University of Bonn, Germany) and their sub-bands (created with the help of wavelet decomposition) are extracted based on it. These parameters are used as a feature to classify the EEG signals into Healthy, Interictal (seizure free) and Ictal (during a seizure) groups using multilayer perceptron (MLP) classifier. Correct classification scores at the range of 91% to 96% reveals the potential of our approach... 

The majority of studies in the field of developing identification and authentication protocols for Internet of Things (IoT) used cryptographic algorithms. Using brain signals is also a relatively new approach in this field. EEG signal-based authentication algorithms typically use feature extraction algorithms that require high processing time. On the other hand, the dynamic nature of the EEG signal makes its use for identification/authentication difficult without relying on feature extraction. This paper presents an EEG-and fingerprint-based two-stage identification-authentication protocol for remote healthcare, which is fast, robust, and multilayer-based. A modified Euclidean distance... 

The majority of studies in the field of developing identification and authentication protocols for Internet of Things (IoT) used cryptographic algorithms. Using brain signals is also a relatively new approach in this field. EEG signal-based authentication algorithms typically use feature extraction algorithms that require high processing time. On the other hand, the dynamic nature of the EEG signal makes its use for identification/authentication difficult without relying on feature extraction. This paper presents an EEG-and fingerprint-based two-stage identification-authentication protocol for remote healthcare, which is fast, robust, and multilayer-based. A modified Euclidean distance... 

In this paper we study the effectiveness of using multiple classifier combination for EEG signal classification aiming to obtain more accurate results than it possible from each of the constituent classifiers. The developed system employs two linear classifiers (SVM,LDA) fused at the abstract and measurement levels for integrating information to reach a collective decision. For making decision, the majority voting scheme has been used. While at the measurement level, two types of combination methods have been investigated: one used fixed combination rules that don't require prior training and a trainable combination method. For the second type, the fuzzy integral method was used. The... 

Steady-State Visual Evoked Potentials (SSVEPs) are one of the most important EEG signals used in Human Computer Interface (HCI) systems. These signals are generated by Looking at flickering external light sources stimulating the central part of the retina. By increasing the number of external light sources, detection of the corresponding SSVEPs from the recorded EEG signal becomes more complicated. On the other hand, the ratio of the sensitivity to specificity in high-speed classifiers becomes more significant. This study presents the effect of the twinkling frequencies and the inter-sources distance of two Light Emitting Diodes (LEDs) on the ratio of the sensitivity to specificity of the... 

Various EEG features have been used in depth of anesthesia (DOA) studies. The objective of this study was to And the excellent features or combination of them than can discriminate between different anesthesia states. Conducting a clinical study on 22 patients we could define 4 distinct anesthetic states: awake, moderate, general anesthesia, and isoelectric. We examined features that have been used in earlier studies using single-channel EEG signal processing method. The maximum accuracy (99.02%) achieved using approximate entropy as the feature. Some other features could well discriminate a particular state of anesthesia. We could completely classify the patterns by means of 3 features and...