Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Saeideh Davoodi; Mohammad Reza Daliri
Volume 11, Issue 3 , September 2017, , Pages 265-273
Abstract
Variety of brain region function represent that interactions between different frequency bands, employ general mechanisms of neural communications. Moreover, a method which recently used for information encoding in the brain is phase synchronization that is a process by which two or more cyclic signals ...
Read More
Variety of brain region function represent that interactions between different frequency bands, employ general mechanisms of neural communications. Moreover, a method which recently used for information encoding in the brain is phase synchronization that is a process by which two or more cyclic signals tends to oscillate with a repeating sequence of relative phase angle. Some evidence demonstrated the important role of phase synchronization in cognitive tasks. In this paper we investigated the role of phase synchronization in a new visual discrimination task. For this purpose we collected electroencephalography signals from fifteen subjects during a color discrimination task. The machine learning algorithm, support vector machine (SVM), was used to find out whether this criterion can distinguish two different colors in the mentioned task. The results show that classification accuracy of 75% is achieved using phase synchronization feature. Also efficient frequency bands and contribution of effective electrodes were shown.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Farin Kahroba; Maryam Mohebbi; Hamed Danandeh Hesar
Volume 11, Issue 2 , June 2017, , Pages 187-199
Abstract
Sudden cardiac death (SCD) is one of the most significant and common causes of heart related deaths around the world. It is believed that SCD can be predicted using signatures and features extracted from ECG signal. These signatures may be seen as arrhythmia or abnormalities in the ECG signal. In this ...
Read More
Sudden cardiac death (SCD) is one of the most significant and common causes of heart related deaths around the world. It is believed that SCD can be predicted using signatures and features extracted from ECG signal. These signatures may be seen as arrhythmia or abnormalities in the ECG signal. In this paper, a monitoring index is introduced for early detection of SCD. This index is acquired by filtering the ECG signal using a nonlinear ECG dynamical model and extended Kalman filter (EKF). The nonlinear dynamical model was a modified version of polar ECG dynamical model proposed by Mc. Sharry et.al. In our algorithm, first the ECG dynamical model is extracted. Then an EKF is applied on the signal. Using the fidelity index extracted from the innovation signal yielded by EKF, a novel algorithm detects the SCD related arrhythmias and abnormalities. The proposed method was evaluated on Physionet Sudden Cardiac Death Holter database. Twenty records corresponding to patients having SCD and eighteen records corresponding to healthy patients were extracted from this database. The evaluation results showed that our proposed monitoring index correctly detected 17 SCDs out of 20 (85% accuracy).
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Alireza Talesh Jafadideh; Babak Mohammadzadeh Asl
Volume 10, Issue 4 , January 2017, , Pages 347-359
Abstract
Minimum variance beamformer (MVB) and its extensions are most widely used techniques in brain source localization due to their high spatial resolution. Unfortunately, beacause of using data covariance matrix, these methods often fail when the number of samples of the recorded data sequences is ...
Read More
Minimum variance beamformer (MVB) and its extensions are most widely used techniques in brain source localization due to their high spatial resolution. Unfortunately, beacause of using data covariance matrix, these methods often fail when the number of samples of the recorded data sequences is small in comparison to the number of electrodes. This condition is particularly relevant when measuring evoked potentials. For solving this problem, Fast Fully Adaptive (FFA) algorithm was developed a few years ago. This method is a multistage adaptive processing technique drawing its inspiration from the butterfly structure of the Fast Fourier Transform (FFT) and decreasing the data requirement significantly. Unfortunately, the high sensitivity of FFA to data partitioning sequences and also its low performance in low SNRs pose a doubt on using it as a reliable localizer for short time brain activities. In this paper, a preprocessing step is proposed to enhance the FFA method. In this step, the brain is divided into separate areas, the components of each area are determined, the data is projected to each area using components of that area. After that, FFA is applied to the projected data. The performance of the enhanced FFA is compared with FFA method by using simulated ERP and real ERF data. In all simulations, enhanced FFA shows the better performance in terms of localization error (enhancement about 2-10 mm) and spread radius (enhancement about 4-9 mm). In addition, the proposed method for real ERF data shows accurate localization result with the most concentrated power spectrum, compared to FFA approach. It is noteworthy that enhanced FFA offers less sensitivity to data partitioning sequences. Emprical results illustrate that enhanced FFA can be implemented as a reliable method for localizing brain short time activities.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Zahra Tabanfar; Seyed Mohammad Firouzabadi; Zeynab Shankaei; Giv Sharifi; Kambiz Novin; Anahita Zoghi
Volume 10, Issue 3 , October 2016, , Pages 211-221
Abstract
In this research, we analyzed the EEG signals of patients with brain tumor and healthy participants in order to study the effects of brain tumor on brain signals and also the feasibility of brain tumor detection using EEG signals. For this reason, EEG signals of four channel F3, F4, T3 and T4 from 5 ...
Read More
In this research, we analyzed the EEG signals of patients with brain tumor and healthy participants in order to study the effects of brain tumor on brain signals and also the feasibility of brain tumor detection using EEG signals. For this reason, EEG signals of four channel F3, F4, T3 and T4 from 5 patients with brain tumor and 4 healthy participants were recorded. After preprocessing, linear features in time and frequency domains and nonlinear ones such as fractal dimensions and entropies were extracted. Afterwards, the differentiation between2 groups was analyzed using Davies-Bouldin Index, LDA, KNN and SVM classifiers. According to the results of Davies-Bouldin Index, RMS, Theta Absolute Power, Approximate Entropy and Sample Entropy features in resting state with eyes closed and RMS and Theta Absolute Power features in resting state with eyes opened, had the most distinction between the two groups. In this stage classification of two groups using single features was done and the most accuracy of 88.89% was obtained for RMS feature in resting state with eyes closed. At the end, classification of two groups using all selected features was conducted and the maximum accuracy of 82.54% was obtained for RMS, Theta Absolute Power, Approximate Entropy and Sample Entropy features in resting state with eyes closed. According to the results, EEG linear features have a good capability of detecting brain tumor. As these features are simple and have low computational complexity, they can be used in online applications especially for periodic screening tests.
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
Masoumeh Rahimi; Mohammad Hasan Moradi; Farnaz Ghassemi
Volume 10, Issue 1 , May 2016, , Pages 59-68
Abstract
The aim of this paper is to study brain effective connectivity based on directed transform function (DTF) using granger causality method. This connectivity was calculated for recorded data in different states of attention and consciousness, forming four different classes: attention-consciousness, attention-unconsciousness, ...
Read More
The aim of this paper is to study brain effective connectivity based on directed transform function (DTF) using granger causality method. This connectivity was calculated for recorded data in different states of attention and consciousness, forming four different classes: attention-consciousness, attention-unconsciousness, inattention-consciousness, and inattention-unconsciousness. Some common indices were extracted and calculated from the connectivity matrices. Indices of these four classes were compared to see whether there is a significant difference among them or not. The Multivariate Autoregressive (MVAR) model was used to obtain the linear causal relations between channels. Furthermore, signals were divided into four frequency bands for more accurate investigation, and the existence of significant difference was investigated with two-way repeated measures test. Results indicated that and among twelve indices could show a significant difference (p<0.05) in five states out of six possible states. The only state that no feature was able to show a meaningful difference was inattention-consciousness, and inattention-unconsciousness.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Alireza Rezaei; Sara Belbasi
Volume 10, Issue 1 , May 2016, , Pages 69-83
Abstract
In this paper, a hybrid algorithm has been developed by analyzing the audio signals of the heart, that consists of extracting features based on chaos technique, reducing dimensions and analyzing the main components and classifying outputs by relying on comparative neuro-fuzzy networks. Uncertainty and ...
Read More
In this paper, a hybrid algorithm has been developed by analyzing the audio signals of the heart, that consists of extracting features based on chaos technique, reducing dimensions and analyzing the main components and classifying outputs by relying on comparative neuro-fuzzy networks. Uncertainty and high error in the diagnosis of inter-ventricular openings are one of the common problems with the previous methods. Due to the importance of the auto-diagnosis of this heart condition, it is necessary to be well-designed and far from error. Transmission of feature spaces to their mapping by the main component analysis algorithm is made by two steps, selecting the number of 18 to 25 attributes among about 50 extracted attributes that these informations are input of the class. The proposed classification classifies the adaptive fuzzy neural network system with the possibility of predicting the incidence of heart disease, which predicts the number of repetitions at the acceptable level of outputs by entering the data. The data are from the Umich database at the University of Michigan and include samples from the ventricular aperture. The ratio of data split in the learning and testing phase is from 0.9 to 0.1 (cross-check), and the K-fold validation method is used. Calculation of criteria such as accuracy, sensitivity and uncertainty by the concept of entropy in a hybrid algorithm suggests the proper performance of the proposed method.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Sara Mohammadi; Ghasem Azemi
Volume 9, Issue 3 , December 2015, , Pages 215-227
Abstract
One of the most important newborn EEG abnormalities is the synchrony between different channels which, according to the clinical studies, can lead to neurological and neurodevelopmental outcomes in adulthood. This paper introduces a new method for automated detection of phase synchrony in multivariate ...
Read More
One of the most important newborn EEG abnormalities is the synchrony between different channels which, according to the clinical studies, can lead to neurological and neurodevelopmental outcomes in adulthood. This paper introduces a new method for automated detection of phase synchrony in multivariate signals with applications to newborn EEG signals. In this method, first the instantaneous phase of each channel of the signal is estimated using Hilbert transform. In the case of EEG signals, due to their multicomponent nature, single-band signalsof the signal are needed to be extracted using a bank of band-pass filters. The synchronization between different channels of the signal is then quantitatively measured using a criterion based on the mutual information between instantaneous phases of theextracted single-band signals. The proposed method in this paper is then used to analyze, from synchronization point of view, multichannel EEG signals acquired from 5 newborns which include seizure-nonseizure periods and burst-suppression (B-S) patterns.Reciever operating curves (ROCs) are used to illustrate the performance of the method. The performance of the proposed method is also compared with that of the existing one based on the cointegration concept. Experimental results prove that the proposed method outperforms the existing one in measuring the generalized phase synchrony in multichannel newborn EEG signals. Also, results of analyzing seizure and nonseizure segments show that for all segmants there is a phase synchronization among EEG channels which is due to the connections between brain hemispheres in both cases. The results also show that seizure periods are more synchronous than nonseizure periods. The phase synchrony assessment of B-S patterns indicates that burst patterns are more synchronous than suppression patterns and there is a phase synchrony in both cases.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mahdi Zolfagharzadeh Kermani; Mohammad Ali Khalilzadeh; Majid Ghoshuni; Peyman Hashemian
Volume 9, Issue 3 , December 2015, , Pages 243-251
Abstract
Evaluation and measurement of parameters associated with methamphetamine craving can be a valuable tool in the management and intervention programs related to methamphetamine use and dependence. We believe that quantitative electroencephalography (EEG) have brought about a revolution in identification ...
Read More
Evaluation and measurement of parameters associated with methamphetamine craving can be a valuable tool in the management and intervention programs related to methamphetamine use and dependence. We believe that quantitative electroencephalography (EEG) have brought about a revolution in identification the neurologic infrastructure of craving processing. This study has been conducted aimed to design and develop a new method to measure baseline craving in methamphetamine-dependent patients using EEG signals in neurofeedback therapy for separation of the three modes of low, medium, and high craving. For this purpose, 10 methamphetamine abusers were selected by available method in terms of age, sex and IQ. All patients received 10 sessions of neurofeedback therapy with alpha-theta protocol. During the period of treatment with neurofeedback, before and 60 minutes after each training session, at rest state, on Pz, for 2 minutes and 10 seconds EEG was recorded. To labeling EEG signals we have used Desire for Drug Questionnaire (DDQ). After collecting the required data from signals, time, frequency and nonlinear features were extracted. Then by calculating the linear correlation coefficient of the two variables and variance analysis on three levels optimized and effective features were selected. Finally, using fuzzy classifier, those features were separated into three classes of low, medium and high craving. According to the results, separation accuracy of EEG signals in three classes by K-fold Cross-Validation (KCV)method was 96.67% and test data was 75.15%. This study showed in addition to estimating baseline craving in methamphetamine abusers by quantifying EEG we can optimize the number of training sessions for participants.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mina Hemmatian; Ali Maleki
Volume 9, Issue 2 , July 2015, , Pages 163-178
Abstract
The humans’ heart is a chaotic system so use of fractal dimension to identify cardiac arrhythmias has been considered. Cardiac arrhythmias are prevalent diseases that is very important to be diagnosed. Hurst index which is calculated using rescaled range analysis method, is used as a criterion ...
Read More
The humans’ heart is a chaotic system so use of fractal dimension to identify cardiac arrhythmias has been considered. Cardiac arrhythmias are prevalent diseases that is very important to be diagnosed. Hurst index which is calculated using rescaled range analysis method, is used as a criterion to evaluate chaotic systems and to quantify the fractal dimensions. Previous studies have shown that classical Hurst index is not appropriate for classification of cardiac arrhythmias because not only selection of algorithm parameters affect the value of determined Hurst index, but also it significantly varies as the heart rate changes. In this paper, modified multiple Hurst index has been proposed to classify the cardiac arrhythmias. The presented index is resistant against changes in heart rate and can be used to identify appropriate features to classify the cardiac arrhythmias. 80 signal from four types of ECG beats obtained from the MIT-BIH Arrhythmia dataset has been used to validate the algorithm. Results show that this method is able to detect normal rhythm and right bundle branch block (RBBB), left bundle branch block (LBBB) and atrial premature complex (APC) arrhythmias with accuracy of 100%, 96.25% and 88.75% using artificialneural network, k nearest neighbor and LDA classifiers respectively.
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
Saleh Lashkari; Mohammad Ali Khalilzadeh; Seyed Mohammad Reza Hashemi Golpayegani
Volume 9, Issue 1 , April 2015, , Pages 59-69
Abstract
Using methods based on nonlinear dynamics such as Poincare Section, can be useful in detecting dynamic biological systems. Selecting a suitable Poincare surface is a critical step in data analysis. Often finding an appropriate position for Poincare section needs to set different parameters. When the ...
Read More
Using methods based on nonlinear dynamics such as Poincare Section, can be useful in detecting dynamic biological systems. Selecting a suitable Poincare surface is a critical step in data analysis. Often finding an appropriate position for Poincare section needs to set different parameters. When the geometry of Poincare surface picks the information related to the stretching and folding, a better discrimination can be performed for the system states. The objective of this paper is to study the effect of position and degree of Poincare surface in Epileptic Seizure Detection. The Poincare surface resulting in the best classification is selected as the optimal section. Accordingly, the phase space of the EEG Segments Reconstructed in three dimension, firstly. Then, a set of Poincare surfaces with 400 different conditions of degree selected to cut the trajectory and Geometric Features Extracted from the points of intersection on each surface. Afterward, extracted features from the Poincare section are applied to SVM classifier. Pearson correlation analysis was performed to analyze the relationship between the classification performance and degree of Poincare section. Certain behavior can be observed by increasing the Surface degree in output classifier. In this way, the increasing and then decreasing pattern were observed by increasing the Surface degree in two Directions of Surface. The results showed that the equation of optimal Poincare Section for m=12 and n=6 gives the accuracy of 96.6%.
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
Mahdi Khezri; Seyed Mohammad Firoozabadi; Seyed Ahmad Reza Sharafat
Volume 8, Issue 4 , February 2015, , Pages 339-358
Abstract
In this study, we propose decision level fusion of multimodal physiological signals to design an affect identification system using the MIT database. Four types of physiological signals, including blood volume pressure (BVP), respiration rate (RSP), skin conductance and facial muscles activities (fEMG) ...
Read More
In this study, we propose decision level fusion of multimodal physiological signals to design an affect identification system using the MIT database. Four types of physiological signals, including blood volume pressure (BVP), respiration rate (RSP), skin conductance and facial muscles activities (fEMG) were utilized as affective modalities. To collect the above-mentioned database, researchers used personalized imagery to elicit the desired affective states from a single subject and recorded the corresponding physiological signals simultaneously. In this study, the best subset of features for each signal was determined using previously calculated time and frequency domain features. To this end, sequential floating forward selection (SFFS) and RELIEF feature selection algorithms were evaluated. A new feature set, formed by concatenating the selected features, was partitioned into three subsets. Each subset was then fed into a classifier to identify the desired affective states. The majority voting method was applied to fuse the results obtained by the subsystems. Three types of classification methods, namely SVM, LDA and KNN were evaluated to design an affect identification system. The results showed remarkable performance from the system in identifying the desired scenarios with an acceptable accuracy and speed of response. Using the RELIEF feature selection method, along with SVM as a classifier, an overall recognition accuracy of 93.8% was obtained, which is better than the results reported with the use of the above-mentioned database so far.
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 Signal Processing / Medical Signal Processing / Biosignal Processing
Afarin Nazemi; Ali Maleki
Volume 8, Issue 4 , February 2015, , Pages 411-420
Abstract
Classification of distal limb movements based on surface electromyography (sEMG) of proximal muscles is an important issue in the control of myoelectric hand prosthesis. In most of previous studies, classification of a limited number of hand motions is investigated. In this paper, we have used NINAPRO ...
Read More
Classification of distal limb movements based on surface electromyography (sEMG) of proximal muscles is an important issue in the control of myoelectric hand prosthesis. In most of previous studies, classification of a limited number of hand motions is investigated. In this paper, we have used NINAPRO database containing kinematics and sEMG of upper limbs while performing 52 finger, hand and wrist movements. We evaluated performance of LDA and LS-SVM with RBF kernel classifiers using different combination of features. First by windowing the signal with two different methods, the major part of the signal was selected and eight various temporal features (MAV, IAV, RMS, WL, E, ER1, ER2, CC) were extracted. Then performance of each classifier with single, double and multiple combinations of features was evaluated. For LDA classifier, the best average classification accuracy of 84.23% was achived for first windowing method and MAV (or IAV)+CC features, The corresponding accuracy for LS-SVM classifier with second windowing method and IAV+MAV+RMS+WL features, was 85.19%.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Sanaz Ahmadzadeh; Hamid Reza Kobravi; Saeed Tosizadeh
Volume 8, Issue 3 , September 2014, , Pages 293-304
Abstract
Multiple muscle groups may be activated simultaneously during the most of activities. So, the appropriate muscle coordination must be emerged during a normal activity. Consequaently, for rehabilitation of movements such as hand writing and paiting in patients for example suffering from carpal channel ...
Read More
Multiple muscle groups may be activated simultaneously during the most of activities. So, the appropriate muscle coordination must be emerged during a normal activity. Consequaently, for rehabilitation of movements such as hand writing and paiting in patients for example suffering from carpal channel syndrom or incomplete spinal cord injury, the correct muscle coordination patterns between the finger muscles and wrist muscles must be reestablished. So, in this paper a prediction methodology based on artificial neural networks (ANN) is proposed to approximate the Thumb fingure extensor and flexor muscles desired activation pattern during the hand writing and Painting. In the presented strategy, A nonlinear auto-regressive neural network (NARX), Recurrent Neural Network (RNN), Radial Basis Function (RBF), Multy Layer Perceptron (MLP) and an Adaptive-network-based fuzzy inference system (ANFIS) are trained to forecast the Extensor pollicis longus and Flexor pollicis brevis muscles activity of one thumb finger of hand using Extensor carpi radialis brevis and Flexor carpi ulnaris muscles activity of forearm. Quantitative evaluations show the promising performance of developed neural networks. Eight healthy volunteers participated in the experiments.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Ali Khadem; Gholam Ali Hossein-Zadeh
Volume 8, Issue 1 , March 2014, , Pages 1-17
Abstract
In EEG/MEG datasets, the Volume Conduction (VC) artifact appears as instantaneous linear mixing of brain source activities on the channel measurements. A desired characteristic of an ideal EEG/MEG connectivity estimator (on sensor-space) is its robustness to VC artifact. This means that the VC of independent ...
Read More
In EEG/MEG datasets, the Volume Conduction (VC) artifact appears as instantaneous linear mixing of brain source activities on the channel measurements. A desired characteristic of an ideal EEG/MEG connectivity estimator (on sensor-space) is its robustness to VC artifact. This means that the VC of independent brain sources must never lead to detection of significant connectivity among EEG/MEG channels. There has been no criterion in the literature so far that can compare the robustness levels of different (sensor-space) connectivity estimators against VC artifact. In this paper, a criterion called Robustness Index (RI) is proposed to compare the robustness levels of connectivity estimators to channel couplings which are modeled by instantaneous linear mixing of quasi-independent components. Since the VC effects have instantaneous linear mixing nature, we expect RI to rank the connectivity estimators according to their robustness levels to VC artifact. RI is used to rank seven functional connectivity estimators: the absolute value of Pearson Correlation Coefficient (CC), Mutual Information (MI), Magnitude Squared Coherence (Coh), (1:1) Phase Locking Value ((1:1)PLV), the absolute value of Imaginary part of Coherency (ImC), Phase Lag Index (PLI) and Weighted Phase Lag Index (WPLI). The results for simulated data and a real EEG dataset show the connectivity estimators that are theoretically robust to VC artifact (ImC, PLI and WPLI) yield RI values near %100 and have the highest ranks, as expected. Also, for the simulated models in which the true VC effects and brain sources are known, ranking the connectivity estimators by RI is consistent with their robustness levels against VC artifact. This supports the possibility of using RI as a tool for ranking the robustness levels of connectivity estimators against VC artifact for real EEG/MEG datasets.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mohsen Naji; Seyed Mohammad Firouzabadi; Sedighe Kahrizi
Volume 7, Issue 1 , June 2013, , Pages 13-20
Abstract
The collected electromyogram (EMG) signals from trunk musculature (e.g., rectus abdominis and external oblique muscle) are often contaminated with the heart muscle electrical activity (ECG). This paper introduces a novel method, the Empirical Mode Decomposition, for elimination of ECG contamination from ...
Read More
The collected electromyogram (EMG) signals from trunk musculature (e.g., rectus abdominis and external oblique muscle) are often contaminated with the heart muscle electrical activity (ECG). This paper introduces a novel method, the Empirical Mode Decomposition, for elimination of ECG contamination from EMG signals. The method is compared to a Butterworth high pass filtering. Results obtained from the analysis of generated and experimental EMG signals show that our method outperforms the high pass filtering for elimination of ECG contamination from trunk EMG signals.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Nasrin Shourie; Seyed Mohammad Firouzabadi; Kambiz Badie
Volume 7, Issue 4 , June 2013, , Pages 321-331
Abstract
In this article, differences between multichannel EEG signals of artists and nonartists were investigated during visual perception and mental imagery of some paintings and at resting condition using scaling exponent. It was found that scaling exponent is significantly higher for artists as compared to ...
Read More
In this article, differences between multichannel EEG signals of artists and nonartists were investigated during visual perception and mental imagery of some paintings and at resting condition using scaling exponent. It was found that scaling exponent is significantly higher for artists as compared to nonartists during the three mentioned states, suggesting that scaling exponent may reflect the influence of artistic expertise. No significant difference in scaling exponent was observed between the visual perception and the mental imagery tasks. In addition, the two groups were classified using scaling exponent of channel C4 and Neural Gas classifier during the visual perception, the mental imagery and the resting condition. The average classification accuracies were 50%, 58.12% and 70%, respectively. The obtained results suggest that discriminability in scaling exponent decreases during the performance of similar cognitive tasks.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mehdi Abdossalehi; Ali Motie Nasrabadi; Seyed Mohammad Firouzabadi
Volume 7, Issue 2 , June 2013, , Pages 143-153
Abstract
In this study, electroencephalogram (EEG) signals have been analyzed in positive, negative and neutral emotions. Here it is supposed that the brain has different independent sources during an emotional activity which will be extractable by Independent Component Analysis (ICA) algorithm. For resolving ...
Read More
In this study, electroencephalogram (EEG) signals have been analyzed in positive, negative and neutral emotions. Here it is supposed that the brain has different independent sources during an emotional activity which will be extractable by Independent Component Analysis (ICA) algorithm. For resolving the illposeness problem of extracted components by ICA algorithm, first these sources were sorted by Shannon entropy and then the features of Katz fractal dimension and the first local minimum of the mutual information based on the time delay (tau) have been extracted for representing determinism. The results show that the determinism ratio of the sorted sources has significant difference during the time in three emotional states: positive, negative and neutral. The determinism ratio increases in neutral, negative and positive emotional states, respectively.
Biomedical Image Processing / Medical Image Processing
Pedram Masaeli; Hamid Behnam; Zahra Alizadeh Sani; Ahmad Shalbaf
Volume 7, Issue 3 , June 2013, , Pages 237-254
Abstract
Coronary artery diseases cause more than half of all deaths in the world. Obviously, early identification is an important way to control coronary artery disease that is diagnosed by measurement and scoring general and regional movement of left ventricle of heart (Normal, Hypokinetic and Akinetic). The ...
Read More
Coronary artery diseases cause more than half of all deaths in the world. Obviously, early identification is an important way to control coronary artery disease that is diagnosed by measurement and scoring general and regional movement of left ventricle of heart (Normal, Hypokinetic and Akinetic). The most common method of imaging the heart using ultrasound is called echocardiography. Using this method accurate view of the heart walls, valves and beginning of main arteries can be obtainbed. Due to the difficulty for the interpretation of these images, time consumption and errors in manual analysis methods, an automated analysis method is required. In this paper we calculate the displacement field in a cycle of heart motion from two-dimensional echocardiography images. To do this, a frame is usually chosen as the reference frame and then all images in a cycle are mapped to it with a mathematical equation. The main idea is to find a semi-local spatiotemporal parametric model for deformation created in a cardiac cycle with nonrigid registration using B-spline functions; as an optimization problem that effectively corrects differences due to movements by minimizing the difference between current frame and a reference frame. Motion estimation accuracy is measured using the sum of squares differences. We use gradient-descend algorithm and multiresolution method to acquire the coefficients in the motion model. The accuracy of the proposed method is assessed using a synthesis sequence of cardiac cycles produced with the simulation software Field II. This algorithm can be applied for the clinical analysis of regional left ventricle then movement parameters and threshold values for the scoring of each section can be extracted. The algorithm represents significant difference between a part of the normal heart and unhealthy heart that shows potential of clinical applications of the proposed method.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Amin Janghorbani; Mohammad Hasan Moradi; Abdollah Arasteh
Volume 7, Issue 2 , June 2013, , Pages 163-174
Abstract
Acute hypotension episodes (AHEs) are one of the hemodynamic instabilities with high mortality rate that is frequent among many groups of patients. Prognosis of acute hypotension episodes can help clinicians to diagnose the cause of this physiological disorder and select proper treatment based on this ...
Read More
Acute hypotension episodes (AHEs) are one of the hemodynamic instabilities with high mortality rate that is frequent among many groups of patients. Prognosis of acute hypotension episodes can help clinicians to diagnose the cause of this physiological disorder and select proper treatment based on this diagnosis. In this study two groups of features, physiological and chaotic features, were extracted from the physiological time series to be applied for prediction of AHEs in the future 1 hour time interval. The best set of the features from the extracted features were selected using Genetic Algorithm (GA) and were classified by SVM. The prediction accuracy for physiological features was 87.5% and for chaotic features was 85%. In order to improve prediction accuracy, physiological and chaotic features were employed simultaneously in feature selection and the best combination of these features was selected by GA and classified by SVM. The best prognosis accuracy, which was achieved in this study by classification of the selected features, was 95% that was better than other previously studies on the same database.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mohsen Mohammadvali’ee; Ali Mahloojifar
Volume 7, Issue 3 , June 2013, , Pages 265-276
Abstract
One of the most important goals for increasing the recognition and treatment revenue is transmitting the vital data to medical care team, more quickly. Nowadays, use of new technologies for transmission of data is extending every day. In this research, for transmitting electrocardiogram, first we code ...
Read More
One of the most important goals for increasing the recognition and treatment revenue is transmitting the vital data to medical care team, more quickly. Nowadays, use of new technologies for transmission of data is extending every day. In this research, for transmitting electrocardiogram, first we code the signal into a matrix of codes, then we will use bluetooth technology to transmit data from offset device to target device. Signal coding will affect in sending and storing data. This suite of codes that form for the first time in this method, include number and type of extermumes, time of occurring them, samples of signal and etc. We complete the coding, using arithmetic coding. The input of arithmetic coding is the extracted suite of coefficients and the output is arithmetic codes. We use SD-200 serial bluetooth modules produced by SENA™ in transmission of coding coefficients. The transmitter sends extracted coefficients and receptor receives them and reconstructs the primary signal. For testing and evaluating the method, we use MIT–BIH arrhythmia database. In our method, when average Percentage of Root Mean Square Differential (PRD) is equal to 5.93%, Compression Ratio (CR) and Cross Correlation (CC) is equal to 8.69 and 99.8%, respectively. Beside, when PRD is about 10.21%, CR and CC is 13.03 and 99.47%, respectively. The maximum standard deviation of compression ratio in two states is 4.17.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mina Amiri; Edmond Zahedi; Fereydoun Behnia
Volume 7, Issue 1 , June 2013, , Pages 85-95
Abstract
It is proved that the endothelial (artery inner lumen cells) function is associated with cardiovascular risk factors. Among all the common non-invasive methods employed in the research setting for assessing endothelial function, flow-mediated dilation is the most widely used one. This technique measures ...
Read More
It is proved that the endothelial (artery inner lumen cells) function is associated with cardiovascular risk factors. Among all the common non-invasive methods employed in the research setting for assessing endothelial function, flow-mediated dilation is the most widely used one. This technique measures endothelial function by inducing reactive hyperemia using temporary arterial occlusion and measuring the resultant relative increase in blood vessel diameter via ultrasound. In this paper, the limitations associated with the ultrasound technique are overcome by using the photoplethysmogram (PPG) signal recorded during FMD. The correctness of this approach is investigated by modeling the AC changes of PPG after FMD by a 2nd order autoregressive model. A sensitivity of 78.6%, specificity of 81.6% and total accuracy of 80% were achieved in classification of 16 healthy and 14 diabetic subjects.