Montri Phothisonothai

EEG-Based Classification of Motor Imagery Tasks Using Fractal Dimension and Neural Network for Brain-Computer Interface

In this study, we propose a method of classifying a spontaneous electroencephalogram (EEG) approach to a brain-computer interface. Ten subjects, aged 21–32 years, volunteered to imagine left-and right-hand movements. An independent component analysis based on a fixed-point algorithm is used to eliminate the activities found in the EEG signals. We use a fractal dimension value to reveal the embedded potential responses in the human brain. The different fractal dimension values between the relaxing and imaging periods are computed. Featured data is classified by a three-layer feed-forward neural network based on a simple backpropagation algorithm. Two conventional methods, namely, the use of the autoregressive (AR) model and the band power estimation (BPE) as features, and the linear discriminant analysis (LDA) as a classifier, are selected for comparison in this study. Experimental results show that the proposed method is more effective than the conventional methods.

ELECTROMYOGRAPHY (EMG) SIGNAL CLASSIFICATION BASED ON DETRENDED FLUCTUATION ANALYSIS

Electromyography (EMG) signal is a useful signal in various medical and engineering applications. To extract the useful information in the EMG signal, feature extraction method should be performed. The extracted features of the EMG signal are usually calculated based on linear or statistical methods, but the EMG signal exhibits the nonlinear and more complex in the properties. With recent advances in nonlinear analysis we are proposing the study of the EMG signals from upper-limb movements using Detrended Fluctuation Analysis (DFA) method. This study used EMG signals obtained from eight upper-limb movements and five muscle positions as representative EMG signals. The usefulness of the DFA method has been proposed to discriminate the upper-limb movements. Complete comparative studies of an optimal parameter of the DFA method were performed. From the viewpoints of maximum class separability, robustness, and complexity, scaling exponent obtained from the DFA method shows the appropriateness to be used as a feature in the classification of the EMG signal. From the experimental results, an optimal DFA method is obtained under these conditions: the minimum box size is approximately four, the maximum box size is one-tenth of the signal length, the box size increment is based on a power of two, and the quadratic polynomial fits is used in the fitting procedure. Moreover, the classification performance of the DFA method is better than other existing nonlinear methods, including the Higuchis method.

EEG signal classification method based on fractal features and neural network

In this paper, we propose a method to classify electroencephalogram (EEG) signal recorded from left- and right-hand movement imaginations. Three subjects (two males and one female) are volunteered to participate in the experiment. We use a technique of complexity measure based on fractal analysis to reveal feature patterns in the EEG signal. Effective algorithm, namely, detrended fluctuation analysis (DFA) has been selected to estimate embedded fractal dimension (FD) values between relaxing and imaging states of the recorded EEG signal. To show the waveform of FDs, we use a windowing-based method or called time-dependent fractal dimension (TDFD) and the Kullback-Leibler (K-L) divergence. Two feature parameters; K-L divergence and different expected values are proposed to be input variables of the classifier. Finally, featured data are classified by a three-layer feed-forward neural network based on a simple backpropagation algorithm. Experimental results can be considerably applied in a brain-computer interface (BCI) application and show that the proposed method is more effective than the conventional method by improving average classification rates of 87.5% and 88.3% for left- and right-hand movement imagery tasks, respectively.

EEG-Based Classification of New Imagery Tasks Using Three-Layer Feedforward Neural Network Classifier for Brain–Computer Interface

In this paper proposes the classification method of new imagery tasks for simple binary commands approach to a brain–computer interface (BCI). An analysis of imaginary tasks as “yes/no” have been proposed. Since BCI is very helpful technology for the patients who are suffering from severe motor disabilities. The BCI applications can be realized by using an electroencephalogram (EEG) signals recording at the scalp surface through the electrodes. Six healthy subjects (three males and three females), aged 23–30 years, were volunteered to participate in the experiment. During the experiment, 10-questions were used to be stimuli. The feature extraction of the event-related synchronization and event-related desynchronization (ERD/ERS) responses can be determined by the slope coefficient and Euclidian distance (SCED) method. The method uses the three-layer feedforward neural network based on a simple backpropagation algorithm to classify the two feature vectors. The experimental results of the proposed method sh...

A CLASSIFICATION METHOD OF DIFFERENT MOTOR IMAGERY TASKS BASED ON FRACTAL FEATURES FOR BRAIN-MACHINE INTERFACE

The objective of this study is to classify spontaneous electroencephalogram (EEG) signal on the basis of fractal concepts. Four motor imagery tasks (left hand movement, right hand movement, feet movement, and tongue movement) were investigated for each EEG recording session. Ten subjects volunteered to participate in this study. As we known, fractal geometry is a mathematical tool for dealing with complex systems like EEG signal. Therefore, we used the fractal dimension (FD) as feature for the application of brain-machine interface (BMI). Effective algorithm, namely, detrended fluctuation analysis (DFA) has been selected to estimate embedded FD values between relaxing and imaging states of the recorded EEG signal. To show the pattern of FDs, we propose a windowing-based method or also called time-dependent fractal dimension (TDFD) and the Kullback-Leibler (K-L) divergence. The K-L divergence and different expected values are employed as the input parameters of classifier. Finally, featured data are classified by a three-layer feed-forward neural network based on a simple backpropagation algorithm. Experimental results show that the proposed method is more effective than the conventional methods.

An investigation of using SSVEP for EEG-based user authentication system

User authentication system to identify individual by using electroencephalograph (EEG) feature based on steady-state visual evoked potential (SSVEP) has been proposed. Recently, SSVEP has been used as a stimulator due to it plays an important role in the response to various visual stimuli, i.e., flickering rate (F), intensity (I), and duty cycle (D). Moreover, the SSVEPs are practical and useful in research because of its excellent signal-to-noise ratio and relative immunity to artifacts and succeeded in many disciplines. In this paper, therefore, we investigate individual SSVEP corresponding to the different visual stimulation in terms of frequency component analysis in the four principal frequency bands, i.e., delta (0.1-3.5 Hz), theta (4.0-7.5 Hz), alpha (8.0-13.0 Hz), and beta (14.0-30.0 Hz). Subjects were instructed to fixate an LED light source then record associated SSVEP waveform. The variation in displaying of the presentation stimuli during a task was examined thereby demonstrating the high usability, adaptability and flexibility of the visual stimulator and determine the optimal parameters for the subject comfort. The experiment achieved the true acceptance rate of 60% to 100% approximately revealing the potential of proposed system for user classification/identification.

Frequency component analysis of eeg recording on various visual tasks: Steady-state visual evoked potential experiment

Nowadays, steady-state visual evoked potential (SSVEP) is ongoing in many research topics. It also plays an important role in the response to various visual stimuli such as flickering rate (F), intensity (I), and duty cycle (D). The SSVEPs are practical and useful in research because of its excellent signal-to-noise ratio and relative immunity to artifacts. The application of using SSVEP-based system has been widely succeeded in many disciplines. In this paper, we investigate SSVEP corresponding to the different visual stimulation in terms of frequency component analysis in the four principal frequency bands, i.e., delta (0.1-3.5 Hz), theta (4.0-7.5 Hz), alpha (8.0-13.0 Hz), and beta (14.0-30.0 Hz). Subjects were instructed to fixate LED light source then record associated SSVEP waveform. The variation in displaying of the presentation stimuli during a task was examined. Experimental results showed that the major frequency distribution has been found in theta and alpha bands.

Time-frequency analysis of duty cycle changing on steady-state visual evoked potential: EEG recording

The purpose of this paper is to investigate the applicability of duty cycle for steady-state visual evoked potential (SSVEP)-based application. For this, the time-frequency analysis of SSVEP during visual stimuli presentation with different duty cycles was performed. The commercial wireless EEG headset with O1 and O2 channels was utilized to record the associated brain activity. The experimental paradigm, in this experiment, we mainly focus on instantaneously changes of duty cycle of SSVEP in the four principal frequency bands, i.e., theta (3-7 Hz), alpha (8-12 Hz), beta (13-30 Hz), and gamma (30-40 Hz). Peak magnitude, peak frequency, and spectral coherence of SSVEP were measured. Results showed that coherences measured in the theta and alpha bands differed significantly with p <; 0.001. Thus, the present results suggest that the duty cycle is recommended as a feature for potential multicommand SSVEP-based applications.

Fractal Analysis of Surface Electromyography (EMG) Signal for Identify Hand Movements Using Critical Exponent Analysis

Recent advances in non-linear analysis have led to understand the complexity and self-similarity of surface electromyography (sEMG) signal. This research paper examines usage of critical exponent analysis method (CEM), a fractal dimension (FD) estimator, to study properties of the sEMG signal and to use these properties to identify various kinds of hand movements for prosthesis control and human-machine interface. The sEMG signals were recorded from ten healthy subjects with seven hand movements and eight muscle positions. Mean values and coefficient of variations of the FDs for all the experiments show that there are larger variations between hand movement types but there is small variation within hand movement. It also shows that the FD related to the self-affine property for the sEMG signal extracted from different hand activities 1.944~2.667. These results have also been evaluated and displayed as a box plot and analysis-of-variance (p value). It demonstrates that the FD value is suitable for using as an EMG feature extraction to characterize the sEMG signals compared to the commonly and popular sEMG feature, i.e., root mean square (RMS). The results also indicate that the p values of the FDs for six muscle positions was less than 0.0001 while that of the RMS, a candidate feature, ranged between 0.0003-0.1195. The FD that is computed by the CEM can be applied to be used as a feature for different kinds of sEMG application.

A Complexity Measure Based on Modified Zero-Crossing Rate Function for Biomedical Signal Processing

A complexity measure is a mathematical tool for analyzing time-series data in many research fields. Various measures of complexity were developed to compare time series and distinguish whether input time-series data are regular, chaotic, and random behavior. This paper proposes a simple technique to measure fractal dimension (FD) values on the basis of zero-crossing function with detrending technique or is called modified zero-crossing rate (MZCR) function. The conventional method, namely, Higuchi’s method has been selected to compare output accuracies. We used the functional Brownian motion (fBm) signal which can easily change its FD for assessing performances of the proposed method. During experiment, we tested the MZCR-based method to determine the FD values of the EEG signal of motor movements. The obtained results show that the complexity of fBm signal is measured in the form of a negative slope of log-log plot. The Hurst exponent and the FD values can be measured effectively.