AR spectral analysis of EEG signals by using maximum likelihood estimation

The applied principles of EEG analysis methods in neuroscience and clinical neurology

Electroencephalography (EEG) is a non-invasive measurement method for brain activity. Due to its safety, high resolution, and hypersensitivity to dynamic changes in brain neural signals, EEG has aroused much interest in scientific research and medical fields. This article reviews the types of EEG signals, multiple EEG signal analysis methods, and the application of relevant methods in the neuroscience field and for diagnosing neurological diseases. First, three types of EEG signals, including time-invariant EEG, accurate event-related EEG, and random event-related EEG, are introduced. Second, five main directions for the methods of EEG analysis, including power spectrum analysis, time-frequency analysis, connectivity analysis, source localization methods, and machine learning methods, are described in the main section, along with different sub-methods and effect evaluations for solving the same problem. Finally, the application scenarios of different EEG analysis methods are emphasized, and the advantages and disadvantages of similar methods are distinguished. This article is expected to assist researchers in selecting suitable EEG analysis methods based on their research objectives, provide references for subsequent research, and summarize current issues and prospects for the future.

Estimation of Autoregressive Parameters from Noisy Observations Using Iterated Covariance Updates

Estimating the parameters of the autoregressive (AR) random process is a problem that has been well-studied. In many applications, only noisy measurements of AR process are available. The effect of the additive noise is that the system can be modeled as an AR model with colored noise, even when the measurement noise is white, where the correlation matrix depends on the AR parameters. Because of the correlation, it is expedient to compute using multiple stacked observations. Performing a weighted least-squares estimation of the AR parameters using an inverse covariance weighting can provide significantly better parameter estimates, with improvement increasing with the stack depth. The estimation algorithm is essentially a vector RLS adaptive filter, with time-varying covariance matrix. Different ways of estimating the unknown covariance are presented, as well as a method to estimate the variances of the AR and observation noise. The notation is extended to vector autoregressive (VAR) processes. Simulation results demonstrate performance improvements in coefficient error and in spectrum estimation.

Nonparametric collective spectral density estimation with an application to clustering the brain signals

In this paper, we develop a method for the simultaneous estimation of spectral density functions (SDFs) for a collection of stationary time series that share some common features. Due to the similarities among the SDFs, the log-SDF can be represented using a common set of basis functions. The basis shared by the collection of the log-SDFs is estimated as a low-dimensional manifold of a large space spanned by a prespecified rich basis. A collective estimation approach pools information and borrows strength across the SDFs to achieve better estimation efficiency. Moreover, each estimated spectral density has a concise representation using the coefficients of the basis expansion, and these coefficients can be used for visualization, clustering, and classification purposes. The Whittle pseudo-maximum likelihood approach is used to fit the model and an alternating blockwise Newton-type algorithm is developed for the computation. A web-based shiny App found at "https://ncsde.shinyapps.io/NCSDE" is developed for visualization, training, and learning the SDFs collectively using the proposed technique. Finally, we apply our method to cluster similar brain signals recorded by the for identifying synchronized brain regions according to their spectral densities.

n-Iterative Exponential Forgetting Factor for EEG Signals Parameter Estimation

Electroencephalograms (EEG) signals are of interest because of their relationship with physiological activities, allowing a description of motion, speaking, or thinking. Important research has been developed to take advantage of EEG using classification or predictor algorithms based on parameters that help to describe the signal behavior. Thus, great importance should be taken to feature extraction which is complicated for the Parameter Estimation (PE)–System Identification (SI) process. When based on an average approximation, nonstationary characteristics are presented. For PE the comparison of three forms of iterative-recursive uses of the Exponential Forgetting Factor (EFF) combined with a linear function to identify a synthetic stochastic signal is presented. The one with best results seen through the functional error is applied to approximate an EEG signal for a simple classification example, showing the effectiveness of our proposal.

Detection of time-, frequency- and direction-resolved communication within brain networks

Electroencephalography (EEG) records fast-changing neuronal signalling and communication and thus can offer a deep understanding of cognitive processes. However, traditional data analyses which employ the Fast-Fourier Transform (FFT) have been of limited use as they do not allow time- and frequency-resolved tracking of brain activity and detection of directional connectivity. Here, we applied advanced qEEG tools using autoregressive (AR) modelling, alongside traditional approaches, to murine data sets from common research scenarios: (a) the effect of age on resting EEG; (b) drug actions on non-rapid eye movement (NREM) sleep EEG (pharmaco-EEG); and (c) dynamic EEG profiles during correct vs incorrect spontaneous alternation responses in the Y-maze. AR analyses of short data strips reliably detected age- and drug-induced spectral EEG changes, while renormalized partial directed coherence (rPDC) reported direction- and time-resolved connectivity dynamics in mice. Our approach allows for the first time inference of behaviour- and stage-dependent data in a time- and frequency-resolved manner, and offers insights into brain networks that underlie working memory processing beyond what can be achieved with traditional methods.

Epileptic Seizures Prediction Using Machine Learning Methods

Epileptic seizures occur due to disorder in brain functionality which can affect patient's health. Prediction of epileptic seizures before the beginning of the onset is quite useful for preventing the seizure by medication. Machine learning techniques and computational methods are used for predicting epileptic seizures from Electroencephalograms (EEG) signals. However, preprocessing of EEG signals for noise removal and features extraction are two major issues that have an adverse effect on both anticipation time and true positive prediction rate. Therefore, we propose a model that provides reliable methods of both preprocessing and feature extraction. Our model predicts epileptic seizures' sufficient time before the onset of seizure starts and provides a better true positive rate. We have applied empirical mode decomposition (EMD) for preprocessing and have extracted time and frequency domain features for training a prediction model. The proposed model detects the start of the preictal state, which is the state that starts few minutes before the onset of the seizure, with a higher true positive rate compared to traditional methods, 92.23%, and maximum anticipation time of 33 minutes and average prediction time of 23.6 minutes on scalp EEG CHB-MIT dataset of 22 subjects.

Algorithm based on the short-term Rényi entropy and IF estimation for noisy EEG signals analysis

Stochastic electroencephalogram (EEG) signals are known to be nonstationary and often multicomponential. Detecting and extracting their components may help clinicians to localize brain neurological dysfunctionalities for patients with motor control disorders due to the fact that movement-related cortical activities are reflected in spectral EEG changes. A new algorithm for EEG signal components detection from its time-frequency distribution (TFD) has been proposed in this paper. The algorithm utilizes the modification of the Rényi entropy-based technique for number of components estimation, called short-term Rényi entropy (STRE), and upgraded by an iterative algorithm which was shown to enhance existing approaches. Combined with instantaneous frequency (IF) estimation, the proposed method was applied to EEG signal analysis both in noise-free and noisy environments for limb movements EEG signals, and was shown to be an efficient technique providing spectral description of brain activities at each electrode location up to moderate additive noise levels. Furthermore, the obtained information concerning the number of EEG signal components and their IFs show potentials to enhance diagnostics and treatment of neurological disorders for patients with motor control illnesses.

Autoregressive model in the Lp norm space for EEG analysis

The autoregressive (AR) model is widely used in electroencephalogram (EEG) analyses such as waveform fitting, spectrum estimation, and system identification. In real applications, EEGs are inevitably contaminated with unexpected outlier artifacts, and this must be overcome. However, most of the current AR models are based on the L2 norm structure, which exaggerates the outlier effect due to the square property of the L2 norm. In this paper, a novel AR object function is constructed in the Lp (p≤1) norm space with the aim to compress the outlier effects on EEG analysis, and a fast iteration procedure is developed to solve this new AR model. The quantitative evaluation using simulated EEGs with outliers proves that the proposed Lp (p≤1) AR can estimate the AR parameters more robustly than the Yule-Walker, Burg and LS methods, under various simulated outlier conditions. The actual application to the resting EEG recording with ocular artifacts also demonstrates that Lp (p≤1) AR can effectively address the outliers and recover a resting EEG power spectrum that is more consistent with its physiological basis.

Adaptation of motor imagery EEG classification model based on tensor decomposition

OBJECTIVE

Session-to-session nonstationarity is inherent in brain-computer interfaces based on electroencephalography. The objective of this paper is to quantify the mismatch between the training model and test data caused by nonstationarity and to adapt the model towards minimizing the mismatch.

APPROACH

We employ a tensor model to estimate the mismatch in a semi-supervised manner, and the estimate is regularized in the discriminative objective function.

MAIN RESULTS

The performance of the proposed adaptation method was evaluated on a dataset recorded from 16 subjects performing motor imagery tasks on different days. The classification results validated the advantage of the proposed method in comparison with other regularization-based or spatial filter adaptation approaches. Experimental results also showed that there is a significant correlation between the quantified mismatch and the classification accuracy.

SIGNIFICANCE

The proposed method approached the nonstationarity issue from the perspective of data-model mismatch, which is more direct than data variation measurement. The results also demonstrated that the proposed method is effective in enhancing the performance of the feature extraction model.

Discriminative Learning of Propagation and Spatial Pattern for Motor Imagery EEG Analysis

Effective learning and recovery of relevant source brain activity patterns is a major challenge to brain-computer interface using scalp EEG. Various spatial filtering solutions have been developed. Most current methods estimate an instantaneous demixing with the assumption of uncorrelatedness of the source signals. However, recent evidence in neuroscience suggests that multiple brain regions cooperate, especially during motor imagery, a major modality of brain activity for brain-computer interface. In this sense, methods that assume uncorrelatedness of the sources become inaccurate. Therefore, we are promoting a new methodology that considers both volume conduction effect and signal propagation between multiple brain regions. Specifically, we propose a novel discriminative algorithm for joint learning of propagation and spatial pattern with an iterative optimization solution. To validate the new methodology, we conduct experiments involving 16 healthy subjects and perform numerical analysis of the proposed algorithm for EEG classification in motor imagery brain-computer interface. Results from extensive analysis validate the effectiveness of the new methodology with high statistical significance.

A unified approach for detection of induced epileptic seizures in rats using ECoG signals

OBJECTIVE

Epileptic seizure detection is a key step for epilepsy assessment. In this work, using the pentylenetetrazole (PTZ) model, seizures were induced in rats, and ECoG signals in interictal, preictal, ictal, and postictal periods were recorded. The recorded ECoG signals were then analyzed to detect epileptic seizures in the epileptic rats.

METHODS

Two different approaches were considered in this work: thresholding and classification. In the thresholding approach, a feature is calculated in consecutive windows, and the resulted index is tracked over time and compared with a threshold. The moment the index crosses the threshold is considered as the moment of seizure onset. In the classification approach, features are extracted from before, during, and after ictal periods and statistically analyzed. Statistical characteristics of some features have a significant difference among these periods, thus resulting in epileptic seizure detection.

RESULTS

Several features were examined in the thresholding approach. Nonlinear energy and coastline features were successful in epileptic seizure detection. The best result was achieved by the coastline feature, which led to a mean of a 2-second delay in its correct detections. In the classification approach, the best result was achieved using the fuzzy similarity index that led to Pvalue<0.001.

CONCLUSION

This study showed that variance-based features were more appropriate for tracking abrupt changes in ECoG signals. Therefore, these features perform better in seizure onset estimation, whereas nonlinear features or indices, which are based on dynamical systems, can better track the transition of neural system to ictal period.

SIGNIFICANCE

This paper presents examination of different features and indices for detection of induced epileptic seizures from rat's ECoG signals.

AUTOMATIC IDENTIFICATION OF EPILEPTIC AND BACKGROUND EEG SIGNALS USING FREQUENCY DOMAIN PARAMETERS

The analysis of electroencephalograms continues to be a problem due to our limited understanding of the signal origin. This limited understanding leads to ill-defined models, which in turn make it hard to design effective evaluation methods. Despite these shortcomings, electroencephalogram analysis is a valuable tool in the evaluation of neurological disorders and the evaluation of overall cerebral activity. We compared different model based power spectral density estimation methods and different classification methods. Specifically, we used the autoregressive moving average as well as from Yule-Walker and Burg's methods, to extract the power density spectrum from representative signal samples. Local maxima and minima were detected from these spectra. In this paper, the locations of these extrema are used as input to different classifiers. The three classifiers we used were: Gaussian mixture model, artificial neural network, and support vector machine. The classification results are documented with confusion matrices and compared with receiver operating characteristic curves. We found that Burg's method for spectrum estimation together with a support vector machine classifier yields the best classification results. This combination reaches a classification rate of 93.33%, the sensitivity is 98.33% and the specificy is 96.67%.

Analyzing circadian expression data by harmonic regression based on autoregressive spectral estimation

Motivation: Circadian rhythms are prevalent in most organisms. Identification of circadian-regulated genes is a crucial step in discovering underlying pathways and processes that are clock-controlled. Such genes are largely detected by searching periodic patterns in microarray data. However, temporal gene expression profiles usually have a short time-series with low sampling frequency and high levels of noise. This makes circadian rhythmic analysis of temporal microarray data very challenging.

Results: We propose an algorithm named ARSER, which combines time domain and frequency domain analysis for extracting and characterizing rhythmic expression profiles from temporal microarray data. ARSER employs autoregressive spectral estimation to predict an expression profile's periodicity from the frequency spectrum and then models the rhythmic patterns by using a harmonic regression model to fit the time-series. ARSER describes the rhythmic patterns by four parameters: period, phase, amplitude and mean level, and measures the multiple testing significance by false discovery rate q-value. When tested on well defined periodic and non-periodic short time-series data, ARSER was superior to two existing and widely-used methods, COSOPT and Fisher's G-test, during identification of sinusoidal and non-sinusoidal periodic patterns in short, noisy and non-stationary time-series. Finally, analysis of Arabidopsis microarray data using ARSER led to identification of a novel set of previously undetected non-sinusoidal periodic transcripts, which may lead to new insights into molecular mechanisms of circadian rhythms.

Availability: ARSER is implemented by Python and R. All source codes are available from http://bioinformatics.cau.edu.cn/ARSER

Contact:zhensu@cau.edu.cn

Neuronal population oscillations of rat hippocampus during epileptic seizures

Neuronal population oscillations in the hippocampus have an important effect in the information processing in the brain and the generation of epileptic seizures. In this paper, we investigate the neuronal population oscillations in the hippocampus of epileptic rats in vivo using an empirical mode decomposition (EMD) method. A neuronal population oscillation can be decomposed into several relaxation oscillations, which possess a recovery and release phase, with the different frequencies that ranges from 0 to 600 Hz. The natures of relaxation oscillations at the pre-ictal, seizure onset and ictal states are distinctly different. The analysis of relaxation oscillations show that the gamma wave is a lead relaxation oscillation at the pre-ictal stage, then it moves to beta oscillation or theta oscillation while the ictal stage starts; the fast relaxation oscillations are associated with the slow relaxation oscillations in the CA1 or CA3, in particular, the fast relaxation oscillations are associated on the recovery phase of the slow relaxation oscillations during the pre-ictal interval, however move to the release phase of the slow relaxation oscillations during the ictal interval. Comparison of the relaxation oscillations in CA1 and CA3 shows that the neurons in the CA1 are more active during the epileptic seizures than during the pre-ictal stage. These findings demonstrate that this method is very helpful to decompose neuronal population for understanding the underlying mechanism of epileptic seizures.

Selection of optimal AR spectral estimation method for EEG signals using Cramer–Rao bound

Electroencephalography is an essential clinical tool for the evaluation and treatment of neurophysiologic disorders related to epilepsy. Careful analyses of the electroencephalograph (EEG) records can provide valuable insight and improved understanding of the mechanisms causing epileptic disorders. The detection of epileptiform discharges in the EEG is an important element in the diagnosis of epilepsy. In this study, EEG signals recorded from 30 subjects were processed using autoregressive (AR) method and EEG power spectra were obtained. The parameters of autoregressive method were estimated by different methods such as Yule-Walker, covariance, modified covariance, Burg, least squares, and maximum likelihood estimation (MLE). EEG spectra were then used to analyze and characterize epileptiform discharges in the form of 3-Hz spike and wave complexes in patients with absence seizures. The variations in the shape of the EEG power spectra were examined in order to obtain medical information. These power spectra were then used to compare the applied methods in terms of their frequency resolution and determination of epileptic seizure. The Cramer-Rao bounds (CRB) were derived for the estimated AR parameters of the EEG signals and the performance evaluation of the estimation methods was performed using the CRB values. Finally, the optimal AR spectral estimation method for the EEG signals was selected according to the computed CRB values. According to the computed CRB values, the performance characteristics of the MLE AR method was found extremely valuable in EEG signal analysis.

DESIGN AND IMPLEMENTATION OF WIRELESS MULTI-CHANNEL EEG RECORDING SYSTEM AND STUDY OF EEG CLUSTERING METHOD

A multi-channel wireless EEG (electroencephalogram) acquisition and recording system is developed in this work. The system includes an EEG sensing and transmission unit and a digital processing circuit. The former is composed of pre-amplifiers, filters, and gain amplifiers. The kernel of the later digital processing circuit is a micro-controller unit (MCU, TI-MSP430), which is utilized to convert the EEG signals into digital signals and fulfill the digital filtering. By means of Bluetooth communication module, the digitized signals are sent to the back-end such as PC or PDA. Thus, the patient's EEG signal can be observed and stored without any long cables such that the analogue distortion caused by long distance transmission can be reduced significantly. Furthermore, an integrated classification method, consisting of non-linear energy operator (NLEO), autoregressive (AR) model, and bisecting k-means algorithm, is also proposed to perform EEG off-line clustering at the back-end. First, the NLEO algorithm is utilized to divide the EEG signals into many small signal segments according to the features of the amplitude and frequency of EEG signals. The AR model is then applied to extract two characteristic values, i.e., frequency and amplitude (peak to peak value), of each segment and to form characteristic matrix for each segment of EEG signal. Finally, the improved modified k-means algorithm is utilized to assort similar EEG segments into better data classification, which allows accessing the long-term EEG signals more quickly.

Comparison of AR and Welch Methods in Epileptic Seizure Detection

Brain is one of the most critical organs of the body. Synchronous neuronal discharges generate rhythmic potential fluctuations, which can be recorded from the scalp through electroencephalography. The electroencephalogram (EEG) can be roughly defined as the mean electrical activity measured at different sites of the head. EEG patterns correlated with normal functions and diseases of the central nervous system. In this study, EEG signals were analyzed by using autoregressive (parametric) and Welch (non-parametric) spectral estimation methods. The parameters of autoregressive (AR) method were estimated by using Yule-Walker, covariance and modified covariance methods. EEG spectra were then used to compare the applied estimation methods in terms of their frequency resolution and the effects in determination of spectral components. The variations in the shape of the EEG power spectra were examined in order to epileptic seizures detection. Performance of the proposed methods was evaluated by means of power spectral densities (PSDs). Graphical results comparing the performance of the proposed methods with that of Welch technique were given. The results demonstrate consistently superior performance of the covariance methods over Yule-Walker AR and Welch methods.

Classification of EMG signals using wavelet neural network

An accurate and computationally efficient means of classifying electromyographic (EMG) signal patterns has been the subject of considerable research effort in recent years. Quantitative analysis of EMG signals provides an important source of information for the diagnosis of neuromuscular disorders. Following the recent development of computer-aided EMG equipment, different methodologies in the time domain and frequency domain have been followed for quantitative analysis. In this study, feedforward error backpropagation artificial neural networks (FEBANN) and wavelet neural networks (WNN) based classifiers were developed and compared in relation to their accuracy in classification of EMG signals. In these methods, we used an autoregressive (AR) model of EMG signals as an input to classification system. A total of 1200 MUPs obtained from 7 normal subjects, 7 subjects suffering from myopathy and 13 subjects suffering from neurogenic disease were analyzed. The success rate for the WNN technique was 90.7% and for the FEBANN technique 88%. The comparisons between the developed classifiers were primarily based on a number of scalar performance measures pertaining to the classification. The WNN-based classifier outperformed the FEBANN counterpart. The proposed WNN classification may support expert decisions and add weight to EMG differential diagnosis.

Application of adaptive neuro-fuzzy inference system for epileptic seizure detection using wavelet feature extraction

Intelligent computing tools such as artificial neural network (ANN) and fuzzy logic approaches are demonstrated to be competent when applied individually to a variety of problems. Recently, there has been a growing interest in combining both these approaches, and as a result, neuro-fuzzy computing techniques have been evolved. In this study, a new approach based on an adaptive neuro-fuzzy inference system (ANFIS) was presented for epileptic seizure detection. The proposed ANFIS model combined the neural network adaptive capabilities and the fuzzy logic qualitative approach. Decision making was performed in two stages: feature extraction using the wavelet transform (WT) and the ANFIS trained with the backpropagation gradient descent method in combination with the least squares method. Some conclusions concerning the impacts of features on the detection of epileptic seizures were obtained through analysis of the ANFIS. The results are highly promising, and a comparative analysis suggests that the proposed modeling approach outperforms ANN model in terms of training performances and classification accuracies. The results confirmed that the proposed ANFIS model has some potential in epileptic seizure detection. The ANFIS model achieved accuracy rates which were higher than that of the stand-alone neural network model.

Comparison of subspace-based methods with AR parametric methods in epileptic seizure detection

Electroencephalography is an important clinical tool for the evaluation and treatment of neurophysiologic disorders related to epilepsy. Careful analyses of the electroencephalograph (EEG) records can provide valuable insight and improved understanding of the mechanisms causing epileptic disorders. The detection of epileptiform discharges in the EEG is an important component in the diagnosis of epilepsy. In this study, we have proposed subspace-based methods to analyze and characterize epileptiform discharges in the form of 3-Hz spike and wave complex in patients with absence seizure. The variations in the shape of the EEG power spectra were examined in order to obtain medical information. These power spectra were then used to compare the applied methods in terms of their frequency resolution and the effects in determination of epileptic seizure. Global performance of the proposed methods was evaluated by means of the visual inspection of power spectral densities (PSDs). Graphical results comparing the performance of the proposed methods with that of the autoregressive techniques were given. The results demonstrate consistently superior performance of the proposed methods over the autoregressive ones.

Comparison of STFT and wavelet transform methods in determining epileptic seizure activity in EEG signals for real-time application

Electroencephalography (EEG) is widely used in clinical settings to investigate neuropathology. Since EEG signals contain a wealth of information about brain functions, there are many approaches to analyzing EEG signals with spectral techniques. In this study, the short-time Fourier transform (STFT) and wavelet transform (WT) were applied to EEG signals obtained from a normal child and from a child having an epileptic seizure. For this purpose, we developed a program using Labview software. Labview is an application development environment that uses a graphical language G, usable with an online applicable National Instruments data acquisition card. In order to obtain clinically interpretable results, frequency band activities of delta, theta, alpha and beta signals were mapped onto frequency-time axes using the STFT, and 3D WT representations were obtained using the continuous wavelet transform (CWT). Both results were compared, and it was determined that the STFT was more applicable for real-time processing of EEG signals, due to its short process time. However, the CWT still had good resolution and performance high enough for use in clinical and research settings.

Application of Classical and Model-Based Spectral Methods to Describe the State of Alertness in EEG

Electrophysiological recordings are considered a reliable method of assessing a person's alertness. Sleep medicine is asked to offer objective methods to measure daytime alertness, tiredness and sleepiness. In this study, EEG signals recorded from 30 subjects were processed by PC-computer using classical and model-based methods. The classical method (fast Fourier transform) and three model-based methods (Burg autoregresse, moving average, least-squares modified Yule-Walker autoregressive moving average methods) were selected for processing EEG signals to discriminate the alertness level of subject. Power spectra of EEG signals were obtained by using these spectrum analysis techniques. These EEG spectra were then used to compare the applied methods in terms of their frequency resolution and the effects in determination of vigilance state of subject. It is found that, FFT and MA methods have low spectral resolution, these two methods are not appropriate for the analysis of the a wake-sleep correlation. Burg AR and least-squares modified Yule-Walker ARMA methods' performance characteristics have been found extremely valuable for the determination of vigilance state of a healthy subject, because of their clear spectra.

Wavelet neural network classification of EEG signals by using AR model with MLE preprocessing

Since EEG is one of the most important sources of information in therapy of epilepsy, several researchers tried to address the issue of decision support for such a data. In this paper, we introduce two fundamentally different approaches for designing classification models (classifiers); the traditional statistical method based on logistic regression and the emerging computationally powerful techniques based on artificial neural networks (ANNs). Logistic regression as well as feedforward error backpropagation artificial neural networks (FEBANN) and wavelet neural networks (WNN) based classifiers were developed and compared in relation to their accuracy in classification of EEG signals. In these methods we used FFT and autoregressive (AR) model by using maximum likelihood estimation (MLE) of EEG signals as an input to classification system with two discrete outputs: epileptic seizure or nonepileptic seizure. By identifying features in the signal we want to provide an automatic system that will support a physician in the diagnosing process. By applying AR with MLE in connection with WNN, we obtained novel and reliable classifier architecture. The network is constructed by the error backpropagation neural network using Morlet mother wavelet basic function as node activation function. The comparisons between the developed classifiers were primarily based on analysis of the receiver operating characteristic (ROC) curves as well as a number of scalar performance measures pertaining to the classification. The WNN-based classifier outperformed the FEBANN and logistic regression based counterpart. Within the same group, the WNN-based classifier was more accurate than the FEBANN-based classifier, and the logistic regression-based classifier.

Automatic seizure detection in EEG using logistic regression and artificial neural network

The detection of epileptiform discharges in the EEG is an important component in the diagnosis of epilepsy. In this study, multiple signal classification (MUSIC), autoregressive (AR) and periodogram methods were used to get power spectra in patients with absence seizure. The EEG power spectra were used as an input to a classifier. We introduce two fundamentally different approaches for designing classification models (classifiers); the traditional statistical method based on logistic regression (LR) and the emerging computationally powerful techniques based on artificial neural networks (ANNs). LR as well as multilayer perceptron neural network (MLPNN) based classifiers were developed and compared in relation to their accuracy in classification of EEG signals. The comparisons between the developed classifiers were primarily based on analysis of the receiver operating characteristic (ROC) curves as well as a number of scalar performance measures pertaining to the classification. The MLPNN based classifier outperformed the LR based counterpart. Within the same group, the MLPNN-based classifier was more accurate than the LR-based classifier.

Classification of EEG signals using neural network and logistic regression

Epileptic seizures are manifestations of epilepsy. Careful analyses of the electroencephalograph (EEG) records can provide valuable insight and improved understanding of the mechanisms causing epileptic disorders. The detection of epileptiform discharges in the EEG is an important component in the diagnosis of epilepsy. As EEG signals are non-stationary, the conventional method of frequency analysis is not highly successful in diagnostic classification. This paper deals with a novel method of analysis of EEG signals using wavelet transform and classification using artificial neural network (ANN) and logistic regression (LR). Wavelet transform is particularly effective for representing various aspects of non-stationary signals such as trends, discontinuities and repeated patterns where other signal processing approaches fail or are not as effective. Through wavelet decomposition of the EEG records, transient features are accurately captured and localized in both time and frequency context. In epileptic seizure classification we used lifting-based discrete wavelet transform (LBDWT) as a preprocessing method to increase the computational speed. The proposed algorithm reduces the computational load of those algorithms that were based on classical wavelet transform (CWT). In this study, we introduce two fundamentally different approaches for designing classification models (classifiers) the traditional statistical method based on logistic regression and the emerging computationally powerful techniques based on ANN. Logistic regression as well as multilayer perceptron neural network (MLPNN) based classifiers were developed and compared in relation to their accuracy in classification of EEG signals. In these methods we used LBDWT coefficients of EEG signals as an input to classification system with two discrete outputs: epileptic seizure or non-epileptic seizure. By identifying features in the signal we want to provide an automatic system that will support a physician in the diagnosing process. By applying LBDWT in connection with MLPNN, we obtained novel and reliable classifier architecture. The comparisons between the developed classifiers were primarily based on analysis of the receiver operating characteristic (ROC) curves as well as a number of scalar performance measures pertaining to the classification. The MLPNN based classifier outperformed the LR based counterpart. Within the same group, the MLPNN based classifier was more accurate than the LR based classifier.

Neural networks with periodogram and autoregressive spectral analysis methods in detection of epileptic seizure

Approximately 1% of the people in the world suffer from epilepsy. Careful analyses of the electroencephalograph (EEG) records can provide valuable insight and improved understanding of the mechanisms causing epileptic disorders. Predicting the onset of epileptic seizure is an important and difficult biomedical problem, which has attracted substantial attention of the intelligent computing community over the past two decades. The purpose of this work was to investigate the performance of the periodogram and autoregressive (AR) power spectrum methods to extract classifiable features from human electroencephalogram (EEG) by using artificial neural networks (ANN). The feedforward ANN system was trained and tested with the backpropagation algorithm using a large data set of exemplars. We present a method for the automatic comparison of epileptic seizures in EEG, allowing the grouping of seizures having similar overall patterns. Each channel of the EEG is first broken down into segments having relatively stationary characteristics. Features are then calculated for each segment, and all segments of all channels of the seizures of a patient are grouped into clusters of similar morphology. This clustering allows labeling of every EEG segment. Examples from 5 patients with scalp electrodes illustrate the ability of the method to group seizures of similar morphology. It was observed that ANN classification of EEG signals with AR preprocessing gives better results, and these results can also be used for the deduction of epileptic seizure.

Automatic recognition of alertness level by using wavelet transform and artificial neural network

We propose a novel method for automatic recognition of alertness level from full spectrum electroencephalogram (EEG) recordings. This procedure uses power spectral density (PSD) of discrete wavelet transform (DWT) of full spectrum EEG as an input to an artificial neural network (ANN) with three discrete outputs: alert, drowsy and sleep. The error back propagation neural network is selected as a classifier to discriminate the alertness level of a subject. EEG signals were obtained from 30 healthy subjects. The group consisted of 14 females and 16 males with ages ranging from 18 to 65 years and a mean age of 33.5 years, and a body mass index (BMI) of 32.4 +/- 7.3 kg/m2. Alertness level and classification properties of ANN were tested using the data recorded in 12 healthy subjects, whereby the EEG recordings were not used been used to train the ANN. The statistics were used as a measure of potential applicability of the ANN. The accuracy of the ANN was 96 +/- 3% alert, 95 +/- 4% drowsy and 94 +/- 5% sleep. The results suggest that the automatic recognition algorithm is applicable for distinguishing between alert, drowsy and sleep state in recordings that have not been used for the training.

Comparison of FFT and adaptive ARMA methods in transcranial Doppler signals recorded from the cerebral vessels

In this work, transcranial Doppler signals recorded from the temporal region of the brain on 35 patients were transferred to a personal computer by using a 16-bit sound card. Fast Fourier transform and adaptive auto regressive-moving average (A-ARMA) methods were applied to transcranial Doppler frequencies obtained from the middle cerebral artery in the temporal region. Spectral analyses were obtained to compare both methods for medical diagnoses. The sonograms obtained using A-ARMA method give better results for spectral resolution than the FFT method. The sonograms of A-ARMA method offer net envelope and better imaging, so that the determination of blood flow and brain pressure can be calculated more accurately. All diseases show higher resistance to flow than controls with no difference between males and females. Whereas values between disease classes differed, resistance within each class was remarkably constant.