Analyzing EEG signals using the probability estimating guarded neural classifier

This paper introduces a neural network architecture for classifying feature vectors symbolizing portions (or segments) of an electroencephalogram (EEG) trace of a human subject. This classification task is the one that is typically required when developing a so-called brain-computer interface (BCI), which analyzes the EEG signals of a subject in order to "understand" the subject's thoughts. However, instead of merely saying which "category of thoughts" (i.e., which class) the respective input feature vector belongs to, the network described here estimates the probabilities of an EEG segment being associated with each individual class. The network, which is called PeGNC (for probability estimating guarded neural classifier), is tested with two kinds of experiments. In the first experiment, the /spl alpha/-rhythm associated with a human subject closing the eyes is detected online with the help of a frequency-based representation. Since the EEG signal is, in general, always a mixture of numerous action potentials generated simultaneously and it is, thus, very likely that mental activities result in overlapping classes, it is reasonable to believe that the PeGNC network - which does not select any one single class, but determines probability values for each mental category - is particularly suitable for this kind of EEG analysis. The second experiment deals with this issue on the basis of an offline analysis of simulated data.     

Local temporal common spatial patterns for robust single-trial EEG classification.

In this paper, we propose a novel optimal spatio-temporal filter, termed local temporal common spatial patterns (LTCSP), for robust single-trial elctroencephalogram (EEG) classification. Different from classical common spatial patterns (CSP) that uses only global spatial covariances to compute the optimal filter, LTCSP considers temporally local information in the variance modelling. The underlying manifold variances of EEG signals contain more discriminative information. LTCSP is an extension to CSP in the sense that CSP can be derived from LTCSP under a special case. By constructing an adjacency matrix, LTCSP is formulated as an eigenvalue problem. So, LTCSP is computationally as straightforward as CSP. However, LTCSP has better discrimination ability than CSP and is much more robust. Simulated experiment and real EEG classification demonstrate the effectiveness of the proposed LTCSP method.     

Application of tripolar concentric electrodes and prefeature selection algorithm for brain-computer interface.

For persons with severe disabilities, a brain-computer interface (BCI) may be a viable means of communication. Lapalacian electroencephalogram (EEG) has been shown to improve classification in EEG recognition. In this work, the effectiveness of signals from tripolar concentric electrodes and disc electrodes were compared for use as a BCI. Two sets of left/right hand motor imagery EEG signals were acquired. An autoregressive (AR) model was developed for feature extraction with a Mahalanobis distance based linear classifier for classification. An exhaust selection algorithm was employed to analyze three factors before feature extraction. The factors analyzed were 1) length of data in each trial to be used, 2) start position of data, and 3) the order of the AR model. The results showed that tripolar concentric electrodes generated significantly higher classification accuracy than disc electrodes.     

Application of Tripolar Concentric Electrodes and Prefeature Selection Algorithm for Brain–Computer Interface

For persons with severe disabilities, a brain-computer interface (BCI) may be a viable means of communication. Lapalacian electroencephalogram (EEG) has been shown to improve classification in EEG recognition. In this work, the effectiveness of signals from tripolar concentric electrodes and disc electrodes were compared for use as a BCI. Two sets of left/right hand motor imagery EEG signals were acquired. An autoregressive (AR) model was developed for feature extraction with a Mahalanobis distance based linear classifier for classification. An exhaust selection algorithm was employed to analyze three factors before feature extraction. The factors analyzed were 1) length of data in each trial to be used, 2) start position of data, and 3) the order of the AR model. The results showed that tripolar concentric electrodes generated significantly higher classification accuracy than disc electrodes.     

Extraction of Alpha Activities from an EEG Obtained During Gum Chewing

We propose a method for extracting alpha activities from electroencephalogram (EEG) recordings contaminated by noise from sources such as electromyograms (EMGs), electro‐oculograms (EOGs), or blinking, using independent component analysis (ICA) with simple preprocessing. In the preprocessing, Gaussian signals are connected to the back and front of the original EEG data. Then waveforms of the original EEG are not subjected to any change in shape. In this paper, we extract alpha activities from an EEG recorded while a subject chews gum. As a result, the ICA separation accuracy is often improved, and alpha activities during gum chewing can be successfully extracted. Furthermore, we investigate whether or not the extracted signals satisfy the general characteristics of an alpha rhythm. Copyright © 2008 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

生理性精神疲劳EEG信号分析技术

从脑电（EEG）信号分析的角度出发,探讨了传统的生理性精神疲劳的分析方法和现阶段基于EEG信号分析的生理性精神疲劳所采用的新技术、新方法,构建了生理性精神疲劳的分析系统,展望了生理性精神疲劳EEG信号分析技术的发展趋势。     

离散差分模块在癫痫脑电分类中的应用

针对癫痫脑电信号多分类的精度提升问题,提出了一种基于信号转差分模块与卷积模块结合的分类算法.信号转差分模块对原始脑电信号进行多阶差分运算,得到描述其波动特征的差分表示;然后卷积模块动态学习的方式将差分脑电信号转换为图片,利用预训练的卷积神经网络来提取信号特征并实现自动分类.分类结果表明,与现有研究相比,所提出的方法的最...     

薛定谔滤波结合阈值算法在核磁脑电梯度伪迹去噪的应用

基于功能磁共振(fMRI)同步采集的脑电图(EEG)，在使用平均模板相减法(AAS)预处理之后，仍存在梯度残留尖峰伪迹。需要更准确地去除残留尖峰，以减少基于频率的活动推断的干扰，降低时间序列之间的虚假相关性。本文针对EEG数据中尖峰伪迹的特性，先使用薛定谔滤波方法分解并识别包含尖峰的EEG数据，自动减去与EEG幅度相差较大的大部分尖峰成分，然后使用幅度阈值方法，通过逆补余误差定位与EEG幅度相当的残留尖峰，实现对尖峰伪迹的定位与去除。对于模拟信号，该方法得到的信号幅值误差(Er)较薛定谔滤波方法平均提高24.95%，信噪比(SNR)较薛定谔滤波方法提高27.13%；对于真实信号，本文方法得到皮尔逊相关系数明显小于另外4种方法，去噪效果较薛定谔滤波方法提升11.42%。无论是尖峰位于波形波谷，还是高频波动幅度与峰值相当的情况下，薛定谔滤波结合阈值算法较其他方法尖峰识别精度和去噪效果明显提高。此去噪方法为EEG-fMRI的融合研究提供了强有力的支持。     

脑电的非线性检测与双谱分析

本文应用高阶谱技术构成以微机为核心的脑电非线性检测及双谱分析系统。采用非高斯参数模型对脑电信号进行建模,利用模型参数检测脑电信号的非线性特性,分析脑电的参数化双谱估计,分别讨论双谱技术的离散实现和系统硬件设置。     

Linear classification of low-resolution EEG patterns produced by imagined hand movements.

Electroencephalograph (EEG)-based brain-computer interfaces (BCI's) require on-line detection of mental states from spontaneous EEG signals. In this framework, surface Laplacian (SL) transformation of EEG signals has proved to improve the recognition scores of imagined motor activity. The results we obtained in the first year of an European project named adaptive brain interfaces (ABI) suggest that: 1) the detection of mental imagined activity can be obtained by using the signal space projection (SSP) method as a classifier and 2) a particular type of electrodes can be used in such a BCI device, reconciling the benefits of SL waveforms and the need for the use of few electrodes. Recognition of mental activity was attempted on both raw and SL-transformed EEG data from five healthy people performing two mental tasks, namely imagined right and left hand movements.     

A time-series prediction approach for feature extraction in a brain-computer interface.

This paper presents a feature extraction procedure (FEP) for a brain-computer interface (BCI) application where features are extracted from the electroencephalogram (EEG) recorded from subjects performing right and left motor imagery. Two neural networks (NNs) are trained to perform one-step-ahead predictions for the EEG time-series data, where one NN is trained on right motor imagery and the other on left motor imagery. Features are derived from the power (mean squared) of the prediction error or the power of the predicted signals. All features are calculated from a window through which all predicted signals pass. Separability of features is achieved due to the morphological differences of the EEG signals and each NNs specialization to the type of data on which it is trained. Linear discriminant analysis (LDA) is used for classification. This FEP is tested on three subjects off-line and classification accuracy (CA) rates range between 88% and 98%. The approach compares favorably to a well-known adaptive autoregressive (AAR) FEP and also a linear AAR model based prediction approach.     

Preprocessing and time-frequency analysis of newborn EEG seizures

Neurological disease or dysfunction in newborn infants is often first manifested by seizures. Prolonged seizures can result in impaired neurodevelopment or even death. In adults, the clinical signs of seizures are well defined and easily recognized. In newborns, however, the clinical signs are subtle and may be absent or easily missed without constant close observation. This article describes the use of adaptive signal processing techniques for removing artifacts from newborn electroencephalogram (EEG) signals. Three adaptive algorithms have been designed in the context of EEG signals. This preprocessing is necessary before attempting a fine time-frequency analysis of EEG rhythmical activities, such as electrical seizures, corrupted by high amplitude signals. After an overview of newborn EEG signals, the authors describe the data acquisition set-up. They then introduce the basic physiological concepts related to normal and abnormal newborn EEGs and discuss the three adaptive algorithms for artifact removal. They also present time-frequency representations (TFRs) of seizure signals and discuss the estimation and modeling of the instantaneous frequency related to the main ridge of the TFR     

Grael脑电放大器与深度学习的手势实时识别研究

手势识别是人机交互的关键。为了能够更好地实现脑电信号与肌电信号的融合,精准地识别人体的运动,本文建立了一套基于Grael脑电放大器的手势动作实时检测识别的研究系统。通过Grael脑电放大器和Curry8系统采集5个通道的8种不同手势的表面肌电信号(sEMG),并对采集到的sEMG信号进行滤波去噪、滑动窗口分割以及特征提取等预处理的操作;最后采用几种常用的分类器与卷积神经网络(CNN)对不同手势的sEMG信号进行实时分类识别。结果表明CNN的识别准确率最高,能达到92.98%;对每个手势动作进行30次实时识别检测,结果显示识别延迟大概在1～1.5 s,实时识别的精度可高达90%。该系统为将来研究脑电信号与肌电信号的融合提供了一个可行的方法,在人机交互方面展现了巨大的潜力和应用空间。     

Estimating the parameters of neural mass models including time delay and nonlinearity using a particle filter: a preliminary study toward model‐based EEG analysis

Electroencephalogram (EEG ) and local field potential (LFP ) signals are measured for both experimental and clinical purposes which include sleep stage analyses, brain–computer interfaces, and disease diagnosis. EEG and LFP data analyses are typically based on models assuming that the measured data is generated from a biological system and estimate the model parameter values that most accurately reproduce the measured data. Thus, use of a biologically plausible model is important for a model‐based analysis. However, analyses using models that include time delay and nonlinearity have not been reported, even though they are biologically important for EEG generation mechanisms. In this study, we developed a parameter estimation method that uses a particle filter for models with time delay and nonlinearity, which was evaluated with simulations. Simulated EEG data were generated from neural mass models (NMMs ). The NMM parameters were estimated from the generated data. Furthermore, parameters for modeling EEG features of patients with Alzheimer's disease were included in the NMM ; the disease parameters were estimated from the simulated EEG data. We observed that NMM parameters, as well as the disease parameters, were accurately estimated from the simulated data. We conclude that the validity of our method for estimating parameters of NMMs including time delay and nonlinearity is confirmed for simulated EEG data, and these results show the possibility of using our method for model‐based analysis with real EEG data. © 2017 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Improving motor imagery classification with a new BCI design using neuro-fuzzy S-dFasArt

This paper presents an algorithm based on neural networks and fuzzy theory (S-dFasArt) to classify spontaneous mental activities from electroencephalogram (EEG) signals, in order to operate a noninvasive brain-computer interface. The focus is placed on the three-class problem, left-hand movement imagination, right movement imagination and word generation. The algorithm allows a supervised classification of temporal patterns improving the classification rates of the BCI Competition III (Data Set V: multiclass problem, continuous EEG). Using the precomputed data supplied for the competition and following the rules established there, a new method based on S-dFasArt, along with rule prune and voting strategy is proposed. The results have been compared with other published methods improving their success rates.     

基于多域信息融合的脑电情感识别研究

脑电信号识别方法较少将空间、时间和频率信息相融合,为了充分挖掘脑电信号包含的丰富信息,本文提出一种多域信息融合的脑电情感识别方法。该方法利用二维卷积神经网络和一维卷积神经网络相结合的并行卷积神经网络(PCNN)模型学习脑电信号的空间、时间和频率特征,来对人类情感状态进行分类。其中,2D-CNN用于挖掘相邻EEG通道间的空间和频率信息,1D-CNN用于挖掘EEG的时间和频率信息。最后,将两个并行卷积模块提取的信息融合进行情感识别。在数据集SEED上的情感三分类实验结果表明,融合空间、时间、频率特征的PCNN整体分类准确率达到了98.04%,与只提取空频信息的2D-CNN和提取时频信息的1D-CNN相比,准确率分别提高了1.97%和0.60%。并于最近的类似工作相比,本文提出的方法对于脑电情感分类具有一定的优越性。     

Collecting EEG signals in the IMPROVE data library

One of the key issues for the IMPROVE (IMPROVing control of patient status in critical carE) project was to define and build a data library (DL) of annotated data acquired in the intensive-care unit (ICU), with particular reference to problems of mismatch between oxygen utilisation and supply. An additional aim of the IMPROVE study was to test the feasibility and clinical value of including limited monitoring of high-quality long-term EEG signals with the main DL in a restricted number of patients. Such an EEG DL would form a useful basis for testing the applicability and validity of different signal processing and interpretation methods in ICU monitoring, and also demonstrate the degree to which useful information could be obtained by a degree of fusion between systemic and cerebral variables. In this article, we describe the setup for collection of the EEG DL, the tools developed to facilitate visual analysis of the EEG together with simultaneous data from other non-EEG variables, data concerning quality control, and some preliminary observations from detailed visual assessment of EEG patterns in relation to other ICU events     

基于离散小波变换的卷积自编码运动想象脑电信号的分类

左右手运动想象脑电信号(MI-EEG)分类准确率低,制约了相关脑-机接口技术的发展。实验采集了16名健康受试者的运动想象脑电信号,提出了一种基于离散小波变换(DWT)和卷积自编码(CAE)的运动想象脑电信号分类算法。利用离散小波变换将EEG转换成时频矩阵,输入到卷积自编码网络中进行脑电信号的特征分类。该算法在实验数据集和公开数据集上测试都得到了较好的分类结果,静息-想象左手、静息-想象右手、想象左手-想象右手3组EEG在实验数据集上分类准确率分别为97.36%、97.27%、86.82%,在公开数据集上分类准确率分别为99.30%、98.23%、92.67%。离散小波变换结合卷积自编码网络模型在左右手运动想象脑电信号分类应用中比其他深度学习方法(CNN、LSTM、STFT-CNN)性能更优。     

基于小波包变换的癫痫脑电信号特征提取

为了有效识别癫痫脑电信号,提出了一种适合于非平稳脑电信号的特征提取方法。以临床采集的包含癫痫发作期的5组500个EEG公共数据为样本,选择了具有任意多分辨分解特性的小波包变换,对信号进行多尺度分解,并提取了各级节点的小波包系数。将小波包系数能量作为特征值,构建了特征向量并输入到BP神经网络分类器中进行自动识别。实验结果表明,该算法的识别率达到了91.5%。     

A Prior Neurophysiologic Knowledge Free Tensor-Based Scheme for Single Trial EEG Classification

Single trial electroencephalogram (EEG) classification is essential in developing brain-computer interfaces (BCIs). However, popular classification algorithms, e.g., common spatial patterns (CSP), usually highly depend on the prior neurophysiologic knowledge for noise removing, although this knowledge is not always known in practical applications. In this paper, a novel tensor-based scheme is proposed for single trial EEG classification, which performs well without the prior neurophysiologic knowledge. In this scheme, EEG signals are represented in the spatial-spectral-temporal domain by the wavelet transform, the multilinear discriminative subspace is reserved by the general tensor discriminant analysis (GTDA), redundant indiscriminative patterns are removed by Fisher score, and the classification is conducted by the support vector machine (SVM). Applications to three datasets confirm the effectiveness and the robustness of the proposed tensor scheme in analyzing EEG signals, especially in the case of lacking prior neurophysiologic knowledge.