Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mohammad Davood Khalili; Vahid Abootalebi; Hamid Saeedi-Sourck
Volume 16, Issue 1 , May 2022, , Pages 75-94
Abstract
The human brain is one of the most complex and heterogeneous networks, and brain signals contain a lot of information, so researchers in this field are always looking for proper solutions to select meaningful features and reduce the dimension of this information appropriately to lead to better classification. ...
Read More
The human brain is one of the most complex and heterogeneous networks, and brain signals contain a lot of information, so researchers in this field are always looking for proper solutions to select meaningful features and reduce the dimension of this information appropriately to lead to better classification. Two of the new tools for brain signal processing are Graph Signal Processing (GSP) and Meta-heuristic and Evolutionary methods. In this paper, a geometric structure and a mixed structure are considered for the brain graph and the weights of the edges in the mixed structure are calculated by a combination of two measures: geometric distance and correlation. To reduce the graph dimension, the weighted degree metric and a combination of the Kron reduction method and Graph Fourier Transform (KG) are used to properly preserve the information of all vertices of the graph into the selected vertices. Feature extraction is performed by Ledoit-Wolf shrinkage estimation and Tangent Space Mapping (TSM) method. For dimension reduction of extracted features, Principal Component Analysis (PCA) method and feature selection based on Differential Evolution (DE) are used. The selected features are given to several well-known machine learning classifiers. To evaluate the performance of the proposed method, dataset IVa from BCI Competition III has been used. The results show that the average classification accuracy of the proposed KG-PCA method with SVM-RBF and DT classifiers, in the structural graph and the functional-structural graph, is higher than the TSM-GFT method expressed in previous studies, and the DT classifier has achieved an average accuracy of 91.15±1.17. Also, according to the obtained results, the performance of the proposed KG-DE method has been better compared to KG-PCA and in the best case, the average accuracy of the SVM-RBF classifier is equal to 95.50±1.27.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Maryam Tavakoli Najafabadi; Vahid Abootalebi; Farzaneh Shayegh
Volume 10, Issue 1 , May 2016, , Pages 1-10
Abstract
The purpose of this article is to evaluate the efficiency of Canonical Correlation Analysis- Recursive Least Square (CCA-RLS)hybridmethod in ElectroOcluGram (EOG) artifact removal from ElectroEncephaloGram (EEG) signal and compare it with Independent Component Analysis (ICA), Canonical Correlation Analysis ...
Read More
The purpose of this article is to evaluate the efficiency of Canonical Correlation Analysis- Recursive Least Square (CCA-RLS)hybridmethod in ElectroOcluGram (EOG) artifact removal from ElectroEncephaloGram (EEG) signal and compare it with Independent Component Analysis (ICA), Canonical Correlation Analysis (CCA), Recursive Least Square (RLS)methods and ICA-RLS hybrid method. After decomposition of the noisy signal by CCA, the noisy components aredetected based ontheir kurtosis, and are filtered by RLS. As the result,the enhanced signal is reconstructed by mixing the original noise-free components and filtered components. In order to compare the methods quantitatively, two evaluation criteria, namely Mean Square Error (MSE) and Signal to Noise Ratio (SNR) are used.The MSE and SNR average values were calculated for five subject in four different channels. EEG data are taken from BCI2008. According to the results,the combination of CCA-RLS method has better performance compareto the other methods used in this paper.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mohammad Ali Manouchehri; Vahid Abootalebi; Amin Mahnam
Volume 9, Issue 2 , July 2015, , Pages 205-214
Abstract
SSVEP-based BCI systems have attracted attention of many researchers due to their high signal to noise ratio, high information transfer rate and being easy for use. The processing goal of these systems is to detect the stimulus frequency of EEG signal. Among the processing methods for frequency identification ...
Read More
SSVEP-based BCI systems have attracted attention of many researchers due to their high signal to noise ratio, high information transfer rate and being easy for use. The processing goal of these systems is to detect the stimulus frequency of EEG signal. Among the processing methods for frequency identification in SSVEP-based BCI systems, LASSO algorithm has gained great acceptance. Although LASSO has acceptable performance in SSVEP-based BCI systems, it doesn't consider the phase of recorded EEG signal for creating the reference signal. In this paper, the idea of correcting the phase of the reference signal with respect to recorded EEG signal was investigated and a new method called phase corrected LASSO was proposed. For this purpose, first, the optimal EEG channel for frequency identification was determined and then, the performance of the phase corrected LASSO method was compared with standard LASSO method. The results show that the phase corrected LASSO method has better performance compared with the standard LASSO method.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Alireza Mirjalili; Vahid Abootalebi; Mohammad Taghi Sadeghi
Volume 8, Issue 4 , February 2015, , Pages 305-323
Abstract
In recent years, Brain-Computer Interface (BCI) has been noted as a new means of communication between the human brain and his surroundings. In order to set up such a system, the collaboration of several blocks, such as data recording, signal processing and user interface are needed. The signal processing ...
Read More
In recent years, Brain-Computer Interface (BCI) has been noted as a new means of communication between the human brain and his surroundings. In order to set up such a system, the collaboration of several blocks, such as data recording, signal processing and user interface are needed. The signal processing block, includes two units of preprocessing and pattern recognition. Pattern recognition block itself involves two phases: feature extraction and classification. In this paper, the sparse representation based classification (SRC) has been used in the classification block. There are two important issues in using the SRC. These are creating an appropriate dictionary matrix and adopting a proper method for finding the sparse solution for an input data. In this research study, the dictionary matrix is formed by extracting an optimal set of features from the training data. Toward this goal, the common spatial patterns algorithm (CSP) is first used. Sensitivity to noise and the over learning phenomena are the main drawbacks of the CSP algorithm. In order to remove these problems, the regularized common spatial patterns algorithm (RCSP) is employed. In previous studies in within the BCI framework, the standard BP algorithm has been used to find a sparse solution. The main disadvantage of the BP algorithm is that the method is computationally expensive. To overcome this weakness, a recently proposed algorithm namely the SL0 approach is used instead. Our experimental results show that when the number of training samples is limited, the RCSP algorithm outperforms the CSP one. Using the features derived from the RCSP, the average detection rate is in average increased by a factor of 7.53%. Our classification results also show that using the SL0 algorithm, the classification process is highly speeded up as compared to the BP algorithm while an almost equivalent accuracy is achieved.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Fereshte Salimian Rizi; Vahid Abootalebi; Mohammad Taghi Sadeghi
Volume 9, Issue 4 , February 2015, , Pages 387-397
Abstract
Detection of Event Related Potentials (ERP) is an important prerequisite in the ERP-based Brain-Computer Interface (BCI) systems. In order to increase the classification accuracy in these systems, different filtering methods are used for improving the signal to noise ratio. This improvement facilitates ...
Read More
Detection of Event Related Potentials (ERP) is an important prerequisite in the ERP-based Brain-Computer Interface (BCI) systems. In order to increase the classification accuracy in these systems, different filtering methods are used for improving the signal to noise ratio. This improvement facilitates the diagnosis and classification of the ERPs. In a number of studies, the performance of P300 detection systems which are based on common spatial pattern (CSP) and common temporal pattern (CTP) has been investigated. The former uses spatial filters while the latter is based on temporal filters. In these methods the filters are trained such that they maximize variance of one class and simultaneously minimize the other class variance. The associated results show that in P300 speller systems, the temporal filters outperform the spatial filters. In this study, in order to improve the performance of the CTP based systems, a Weighted Common Temporal Pattern (WCTP) algorithm which is a combined method is proposed. In this algorithm, each category of features has a weight based on the importance of its eigenvalues. In fact, the features produced by the initial and final CTP filters have more weight in the decision making process. In the combined method used in this algorithm, the LDA classifiers are used. It is shown that the set of features obtained by the WCTP method leads to an average classification accuracy of 90.2 percent which is about 4 percent better than the CTP method. The experiments are performed considering two different subjects on 5 trials.
Biomedical Image Processing / Medical Image Processing
Malihe Miri; Mohammad Taghi Sadeghi; Vahid Abootalebi
Volume 8, Issue 1 , March 2014, , Pages 45-56
Abstract
Successful outcomes of Sparse Representation-based Classifier (SRC) and Sparse Subspace Clustering (SSC) in many applications motivated us to combine these methods and propose a hierarchical classifier. The main idea behind the SRC and SSC algorithms is to represent a data using a sparse linear combination ...
Read More
Successful outcomes of Sparse Representation-based Classifier (SRC) and Sparse Subspace Clustering (SSC) in many applications motivated us to combine these methods and propose a hierarchical classifier. The main idea behind the SRC and SSC algorithms is to represent a data using a sparse linear combination of elementary signals so that those elementary signals which are similar to the data contribute mainly in the representation. In this paper, the performance of a sparse representation based classifier is improved by pre-clustering of training samples using the SSC algorithm. A twostage SRC is then designed using the resulting clusters. A test data is classified by first determining the most similar cluster. The data label is subsequently found using the second stage classifier. The performance of the proposed method is evaluated considering cancer classification problem using the 14-Tumors microarray dataset. Due to low number of data samples per each class and high dimensionality of the data, this is a challenging problem. Curse of dimensionality, overfitting of the classifier to the training data and computational complexity are the possible related problems. Our experimental results show that the proposed method outperforms some other state of the art classifiers.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mohammad Shahab Shahvazian; Vahid Abootalebi; Mohammad Taghi Sadeghi
Volume 6, Issue 1 , June 2012, , Pages 35-47
Abstract
With the advent of biometric knowledge, conventional methods of authentication are being replaced with biometric based methods. Recently, the use of EEG signal in biometric systems attracted increasing research attention. Only a few works have been done in this emerging of EEG-based biometry mainly focusing ...
Read More
With the advent of biometric knowledge, conventional methods of authentication are being replaced with biometric based methods. Recently, the use of EEG signal in biometric systems attracted increasing research attention. Only a few works have been done in this emerging of EEG-based biometry mainly focusing on person identification not on person authentication. This paper examines the effectiveness of the EEG as a biometric for person authentication. In this study, the EEG signal from fifteen volunteer recorded during imagination of opening and closing fist was used. A set of AR coefficients, power of spectral bands, Energy Spectral Density, Energy Entropy and Sample Entropy were used as extracted features. The authentication system is fused at the sensor module and features to support a system which can meet more challenging and varying requirements. The utility of the sequential search methods is also experimentally studied. In the extensive experimentation on the Shalk and his colleague’s database, we demonstrate that with combination of features when using single channel EEG, the performance of system is improved in two ways of single block and multi block methods compared to other. Result of this study shows a clear vision of commercial and practical use of the brain's electrical signals in the authentication systems of future.
Zahra Amini; Vahid Abootalebi; Mohammad Taghi Sadeghi
Volume 4, Issue 4 , June 2010, , Pages 293-306
Abstract
The aim of this paper is to design a pattern recognition based system to detect P300 component in multi-channel electroencephalogram (EEG) trials. This system has two main blocks, feature extraction and classification. In feature extraction block, in addition to conventional features namely morphological, ...
Read More
The aim of this paper is to design a pattern recognition based system to detect P300 component in multi-channel electroencephalogram (EEG) trials. This system has two main blocks, feature extraction and classification. In feature extraction block, in addition to conventional features namely morphological, frequency and wavelet features, some new features included intelligent segmentation, common spatial pattern (CSP) and combined features (CSP + Segmentation) have also been used. Three criteria were used for evaluation and selection of a feature set by choosing a subset of the original features that contains most of essential information. Firstly, a statistical analysis has been applied for evaluating the fitness of each feature in discriminating between target and non target signals. Secondly, each of these six groups of features was evaluated by a Linear Discriminant Analysis (LDA) classifier. Furthermore by using Stepwise Linear Discriminant Analysis (SWLDA), the best set of features was selected. Among these six feature vectors, intelligent segmentation was seen to be most efficient in classification of these signals. In classification phase, two linear classifiers -LDA and SWLDA- were used. The algorithm was described here has tested with dataset II from the BCI competition 2005. In this research, the best result for P300 detection is 97.05% .This result have proven to be more accurate than the results of previous works carried out in this filed.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Vahid Abootalebi; Mohammad Hasan Moradi; Mohammad Ali Khalilzadeh
Volume -1, Issue 1 , June 2004, , Pages 25-45
Abstract
P300 is the most predominant cognitive component of the brain signals. In this study, the single trial event related potentials recorded from the scalp, were decomposed to their time-frequency components using discrete wavelet transform. These quantities were later analyzed as the features related to ...
Read More
P300 is the most predominant cognitive component of the brain signals. In this study, the single trial event related potentials recorded from the scalp, were decomposed to their time-frequency components using discrete wavelet transform. These quantities were later analyzed as the features related to the cognitive activities of brain. Study on these features showed that cognitive processes of the brain of ten reflected in the feature of δ and θ bands. The aim of this study, as a primary step for "lie detection using brain signals (EEG - Polygraphy)", was to design a system for discriminating between single trials involved P300 and those without it. In the first approach, an optimal discriminant function based on 9 features was designed using "Stepwise Linear Discriminant Analysis". Detection accuracy was 75% in training data and 71% in test data. More study on this method showed that almost similar accuracy could be obtained from the features of Pz channel alone. In the second approach, the modular learning strategy - based on principal component analysis and neural networks - was used. After training the systems, the maximum classification accuracy was 76% in train data and 72% in test data.