Electroencephalogram (EEG) is the primary tool in investigation of the brain science. It is necessary to carry out a deepgoing study into the characteristics and information hidden in EEGs to meet the needs of the clinical research. In this paper, we present a wavelet-nonlinear dynamic methodology for analysis of nonlinear characteristic of EEGs and delta, theta, alpha, and beta sub-bands. We therefore studied the effectiveness of correlation dimension (CD), largest Lyapunov exponen, and approximate entropy (ApEn) in differentiation between the interictal EEG and ictal EEG based on statistical significance of the differences. The results showed that the nonlinear dynamic characteristic of EEG and EEG subbands could be used as effective identification statistics in detecting seizures.
This paper aims to realize the decoding of single trial motor imagery electroencephalogram (EEG) signal by extracting and classifying the optimized features of EEG signal. In the classification and recognition of multi-channel EEG signals, there is often a lack of effective feature selection strategies in the selection of the data of each channel and the dimension of spatial filters. In view of this problem, a method combining sparse idea and greedy search (GS) was proposed to improve the feature extraction of common spatial pattern (CSP). The improved common spatial pattern could effectively overcome the problem of repeated selection of feature patterns in the feature vector space extracted by the traditional method, and make the extracted features have more obvious characteristic differences. Then the extracted features were classified by Fisher linear discriminant analysis (FLDA). The experimental results showed that the classification accuracy obtained by proposed method was 19% higher on average than that of traditional common spatial pattern. And high classification accuracy could be obtained by selecting feature set with small size. The research results obtained in the feature extraction of EEG signals lay the foundation for the realization of motor imagery EEG decoding.
This paper analyzed literatures on the specificity study of electroencephalogram (EEG) in the diagnosis of depression since 2010 to 2020, summarized the recent research directions in this field and prospected the future research hotspots at home and abroad. Based on databases of China National Knowledge Infrastructure (CNKI) and the core collection of Web of Science (WOS), CiteSpace software was used to analyze the relevant literatures in this research field. The number of relevant literatures, countries, authors, research institutions, key words, cited literatures and periodicals related to this research were analyzed, respectively, to explore research hotspots and development trends in this field. A total of 2 155 articles were included in the WOS database. The most published institution was the University of Toronto, the most published country was the United States, China occupied the third place, and the hot keywords were anxiety, disorder, brain and so on. A total of 529 literatures were included and analyzed in CNKI database. The institution with the most publications was the Mental Health Center of West China Hospital of Sichuan University, and the hot keywords were EEG signal, event-related potential, convolutional neural network, schizophrenia, etc. This study finds that EEG study of depression is developing rapidly at home and abroad. Research directions in the world mainly focus on exploring the characteristics of spontaneous EEG rhythm and nonlinear dynamic parameters during sleep in depressed patients. In addition, synchronous transcranial magnetic stimulation (TMS) and EEG technologies also attract much attention abroad, and the future research hotspot will be on the mechanism of EEG on patients with major depression. Domestic research directions mainly focus on the classification of resting EEG and the control study of resting EEG power spectrum entropy in patients with schizophrenia and depression, and future research hotspot is the basic and clinical EEG study of depressed patients complicated with anxiety.
One of the key problems of brain-computer interfaces (BCI) is low signal-to-noise ratio (SNR) of electroencephalogram (EEG) signals. It affects recognition performance. To remove the artifact and noise, block under-determined blind source separation method based on the small number of channels is proposed in this paper. The non-stationary EEG signals are turned into block stationary signals by piecewise. The mixing matrix is estimated by the second-order under-determined blind mixing matrix identification. Then, the beamformer based on minimum mean square error separates the original sources of signals. Eventually, the reconstructed EEG for mixed signals removes the unwanted components of source signals to achieve suppressing artifact. The experiment results on the real motor imagery BCI indicated that the block under-determined blind source separation method could reconstruct signals and remove artifact effectively. The accuracy of motor imagery task of BCI has been greatly improved.
Vigilance is defined as the ability to maintain attention for prolonged periods of time. In order to explore the variation of brain vigilance in work process, we designed addition and subtraction experiment with numbers of three digits to induce the vigilance to change, combined it with psychomotor vigilance task (PVT) to measure this process of electroencephalogram (EEG), extracted and analyzed permutation entropy (PE) of 11 cases of subjects' EEG and made a brief comparison with nonlinear parameter sample entropy (SE). The experimental results showed that:PE could well reflect the dynamic changes of EEG when vigilance decreases, and has advantages of fast arithmetic speed, high noise immunity, and low requirements for EEG length. This can be used as a measure of the brain vigilance indicators.
Electroencephalogram (EEG) analysis has been widely used in disease diagnosis. The EEG detection of the patients with epilepsy can be used to make judgments about patients' conditions in time, which is of great practical value. Therefore, the techniques of automatic detection, diagnosis and classification of epileptic EEG signals are urgently needed. In order to realize fast and accurate automatic detection and classification of the EEG signals during the normal, interictal and ictal periods of epilepsy, we propose an automatic classification and recognition method which combines the Real Adaboost algorithm based on error-correcting output codes (ECOC) with a feature extraction method based on sample entropy (SampEn) and wavelet packet energy in this paper. In the present study, we used the sample entropy of input signals and the energy of some parts of frequency bands as features, and then we classified the extracted features with the method combining ECOC with Real AdaBoost algorithm. In order to test the validity, we used the epilepsy database from the University of Bonn. The database has 5 groups of EEG signals, which contains the data of normal people with their eyes open or closed, the data collected inside and outside of the epileptic foci from patients during their interictal period and the data from patients during their ictal period. The results showed that the method had strong abilities of classification and recognition of the EEG signals, and especially the recognition rate had been improved significantly. The average recognition rate of the EEG signals with different features during the three periods of the five groups mentioned above can reach 96.78%, which is superior to those with algorithms recorded in many other literatures. The method has better stability, processing speed and potential of real-time application, and it plays a supporting role in the prediction and detection of epilepsy in clinical practice.
The incidence of tinnitus is very high, which can affect the patient’s attention, emotion and sleep, and even cause serious psychological distress and suicidal tendency. Currently, there is no uniform and objective method for tinnitus detection and therapy, and the mechanism of tinnitus is still unclear. In this study, we first collected the resting state electroencephalogram (EEG) data of tinnitus patients and healthy subjects. Then the power spectrum topology diagrams were compared of in the band of δ (0.5–3 Hz), θ (4–7 Hz), α (8–13 Hz), β (14–30 Hz) and γ (31–50 Hz) to explore the central mechanism of tinnitus. A total of 16 tinnitus patients and 16 healthy subjects were recruited to participate in the experiment. The results of resting state EEG experiments found that the spectrum power value of tinnitus patients was higher than that of healthy subjects in all concerned frequency bands. The t-test results showed that the significant difference areas were mainly concentrated in the right temporal lobe of the θ and α band, and the temporal lobe, parietal lobe and forehead area of the β and γ band. In addition, we designed an attention-related task experiment to further study the relationship between tinnitus and attention. The results showed that the classification accuracy of tinnitus patients was significantly lower than that of healthy subjects, and the highest classification accuracies were 80.21% and 88.75%, respectively. The experimental results indicate that tinnitus may cause the decrease of patients’ attention.
The functional coupling between motor cortex and effector muscles during autonomic movement can be quantified by calculating the coupling between electroencephalogram (EEG) signal and surface electromyography (sEMG) signal. The maximal information coefficient (MIC) algorithm has been proved to be effective in quantifying the coupling relationship between neural signals, but it also has the problem of time-consuming calculations in actual use. To solve this problem, an improved MIC algorithm was proposed based on the efficient clustering characteristics of K-means ++ algorithm to accurately detect the coupling strength between nonlinear time series. Simulation results showed that the improved MIC algorithm proposed in this paper can capture the coupling relationship between nonlinear time series quickly and accurately under different noise levels. The results of right dorsiflexion experiments in stroke patients showed that the improved method could accurately capture the coupling strength of EEG signal and sEMG signal in the specific frequency band. Compared with the healthy controls, the functional corticomuscular coupling (FCMC) in beta (14~30 Hz) and gamma band (31~45 Hz) were significantly weaker in stroke patients, and the beta-band MIC values were positively correlated with the Fugl-Meyers assessment (FMA) scale scores. The method proposed in this study is hopeful to be a new method for quantitative assessment of motor function for stroke patients.
In this paper, the response of individual's physiological system under psychological stress state is discussed, and the theoretical support for psychological stress assessment research is provided. The two methods, i.e. the psychological stress assessment of questionnaire and physiological parameter assessment used for current psychological stress assessment are summarized. Then, the future trend of development of psychological stress assessment research is pointed out. We hope that this work could do and provide further support and help to psychological stress assessment studies.
As a common disease in nervous system, epilepsy is possessed of characteristics of high incidence, suddenness and recurrent seizures. Timely prediction with corresponding rescues and treatments can be regarded as effective countermeasure to epilepsy emergencies, while most accidental injuries can thus be avoided. Currently, how to use electroencephalogram (EEG) signals to predict seizure is becoming a highlight topic in epilepsy researches. In spite of significant progress that made, more efforts are still to be made before clinical applications. This paper reviews past epilepsy studies, including research records and critical technologies. Contributions of machine learning (ML) and deep learning (DL) on seizure predictions have been emphasized. Since feature selection and model generalization limit prediction ratings of conventional ML measures, DL based seizure predictions predominate future epilepsy studies. Consequently, more exploration may be vitally important for promoting clinical applications of epileptic seizure prediction.
