Chongxun Zheng

Detection of ECG characteristic points using wavelet transforms

An algorithm based on wavelet transforms (WTs) has been developed for detecting ECG characteristic points. With the multiscale feature of WTs, the QRS complex can be distinguished from high P or T waves, noise, baseline drift, and artifacts. The relation between the characteristic points of ECG signal and those of modulus maximum pairs of its WTs is illustrated. By using this method, the detection rate of QRS complexes is above 99.8% for the MIT/BIH database and the P and T waves can also be detected, even with serious baseline drift and noise.

Online Removal of Muscle Artifact from Electroencephalogram Signals Based on Canonical Correlation Analysis

The electroencephalogram (EEG) is often contaminated by electromyography (EMG). In this paper, a novel and robust technique is presented to eliminate EMG artifacts from EEG signals in real-time. First, the canonical correlation analysis (CCA) method is applied on the simulated EEG data contaminated by EMG and electrooculography (EOG) artifacts for separating EMG artifacts from EEG signals. The components responsible for EMG artifacts are distinguished from those responsible for brain activity based on the relative low autocorrelation. We demonstrate that the CCA method is more suitable to reconstruct the EMG-free EEG data than independent component analysis (ICA) methods. In addition, by applying CCA to analyze a number of EEG data contaminated by EMG artifacts, a correlation threshold is determined using an unbiased procedure. Hence, CCA can be used to remove EMG artifacts automatically. Finally, an example is given to verify that, after EMG artifacts were removed successfully from the EEG data contaminated by EMG and EOG simultaneously, not only the underlying brain activity signals but the EOG artifacts are preserved with little distortion.

Bispectrum analysis of focal ischemic cerebral EEG signal using third-order recursion method

In this paper, a model for Sprague-Dawley (SD) rat focal ischemic cerebral injury is presented. Based on this experimental model, the electroencephalogram (EEG) from the ischemic region and from a normal region are collected during the first 30 min of ischemia. The EEG bispectrum analysis is carefully investigated by using the third-order recursion method. The authors found that some characteristics of the bispectrum are very sensitive to focal ischemic cerebral injury. The maximum magnitude and the weighted center of EEG bispectrum (WCOB) change according to the extent and the place of the injury region. The bispectrum analysis results have been verified by the heat shock protein (HSP) test. The study indicates that the EEG bispectrum analysis may be useful to distinguish the ischemic region from the normal one and to estimate the ischemic extent.

Wavelet packet transform for feature extraction of EEG during mental tasks

Wavelet packet transform (WPT) based feature extraction of the electroencephalogram (EEG) is introduced. Six-channel EEG data of four subjects were recorded while they performed three different mental tasks. Approximate one-second data segments were divided and transformed to multi-scale representations by dyadic wavelet packet decomposition channel by channel. Power values of different sub-spaces of six-channel EEG signals formed the feature vectors. A radial basis function (RBF) network was applied to classify the three task pairs. The average classification accuracy of four subjects over three task pairs is 85.3%. Compared with the two autoregressive (AR) model methods, wavelet packet transform would be a promising method to extract features from EEG signals.

Analysis of the Distance Between Two Classes for Tuning SVM Hyperparameters

An important step in the construction of a support vector machine (SVM) is to select optimal hyperparameters. This paper proposes a novel method for tuning the hyperparameters by maximizing the distance between two classes (DBTC) in the feature space. With a normalized kernel function, we find that DBTC can be used as a class separability criterion since the between-class separation and the within-class data distribution are implicitly taken into account. Employing DBTC as an objective function, we develop a gradient-based algorithm to search the optimal kernel parameter. On the basis of the geometric analysis and simulation results, we find that the optimal algorithm and the initialization problem become very simple. Experimental results on the synthetic and real-world data show that the proposed method consistently outperforms other existing hyperparameter tuning methods.

Multivariate autoregressive models and kernel learning algorithms for classifying driving mental fatigue based on electroencephalographic

Long-term driving is a significant cause of fatigue-related accidents. Driving mental fatigue has major implications for transportation system safety. Monitoring physiological signal while driving can provide the possibility to detect the mental fatigue and give the necessary warning. In this paper an EEG-based fatigue countermeasure algorithm is presented to classify the driving mental fatigue. The features of multichannel electroencephalographic (EEG) signals of frontal, central and occipital are extracted by multivariate autoregressive (MVAR) model. Then kernel principal component analysis (KPCA) and support vector machines (SVM) are employed to identify three-class EEG-based driving mental fatigue. The results show that KPCA-SVM method is able to effectively reduce the dimensionality of the feature vectors, speed up the convergence in the training of SVM and achieve higher recognition accuracy (81.64%) of three driving mental fatigue states in 10 subjects. The KPCA-SVM method could be a potential tool for classification of driving mental fatigue.

Automatic recognition of cognitive fatigue from physiological indices by using wavelet packet transform and kernel learning algorithms

Cognitive fatigue is an extremely sophisticated phenomenon, which is influenced by the environment, the state of health, vitality, and the capability of recovery. A single parameter can not fully describe it. In this paper, power spectral indices of HRV and wavelet packet parameters of EEG are firstly combined to analyze the impacts of long time switch task on autonomic nervous system and central nervous system. Then wavelet packet parameters of EEG are extracted as the features of brain activity in different cognitive fatigue state, kernel principal component analysis (KPCA) and support vector machine (SVM) are jointly applied to differentiate two states. The experimental results show that the predominant activity of autonomic nervous system of subjects turns to the sympathetic activity from parasympathetic activity after the task. The wavelet packet parameters of EEG are strongly related with cognitive fatigue. Moreover, the joint KPCA-SVM method is able to effectively reduce the dimensionality of the feature vectors, speed up the convergence in the training of SVM and achieve higher recognition accuracy (90.04%) of cognitive fatigue state. The KPCA-SVM method could be a promising tool for the evaluation of cognitive fatigue.

QRS detection by wavelet transform

In this paper, the wavelet transform is used for QW detection. Some detection strategies are applied to decrease the fake detection succes~fully. The correct QRS detection rate reaches to 99. 8% in the experiment with the bflT/BLH arrhythmia database. INTRODUCTION ECG signal has been extensively studied because of its great clinical significance in diagnosing heart diseases. QRS detection is the most important problem in ECG signal analysis. A number of QRS detectors have been designed, but none of them is perfect. The wavelet transform(WT) is a recently introduced thefrequency Id i za t ion tcdmique which has already found applications in a variety of fields. In essence, the WT in different scale represents the information of signal in different frequency domain. Advantages have been found in its use to locate signal singularities. This paper gives some results in QRS detection with the WT. METHODS I . Wavelet Transform The W T of a signal f (x) is defined as where s is scale; Jl(x) is named a wavelet, Js (x> =~41(:)is the dilation of

Denoised P300 and machine learning-based concealed information test method

(XI by scale factor In order to calculate the WT by computer, the WT should be discretization WT , so lead to a dyadic wavelet transform as Wif (x) = f * @(x), where s= 2’. The dyadic WT can be calculated by Mallat algorithmC31, as described as Spff(n) = x h , S p l f ( n 2J ’ lk ) 1

An efficient automatic framework for segmentation of MRI brain image

In this paper, a novel P300-based concealed information test (CIT) method was proposed to improve the efficiency of differentiating deception and truth-telling. Thirty subjects including the guilty and innocent performed the paradigm based on three types of stimuli. In order to reduce the influence from the occasional variability of cognitive states on the CIT, several single-trials from Pz in probe stimuli within each subject were first averaged. Then the three groups of features were extracted from these averaged single-trials. Finally, two classes of feature samples were used to train a support vector machine (SVM) classifier. Meanwhile, the optimal number of averaged Pz waveforms and some other parameter values in the classifiers were determined by the cross validation procedures. Results show that if choosing accuracy of 90% as a detecting standard of P3 component to classify a subjects status (guilty or innocent), our method can achieve individual diagnostic rate of 100%. The individual diagnostic rate of our method was higher than the results of the other related reports. The presented method improves efficiency of CIT, and is more practical, lower fatigue and less countermeasure behavior in comparison with previous report methods, which could extend the laboratory study to the practical application.