卷积神经网络在脑疲劳检测中的研究

脑疲劳是由于持续进行脑力劳动导致的一种状态，脑电被认为是脑疲劳状态检测的最佳工具。如何选取合适的脑疲劳特征成为脑疲劳检测的关键问题，传统模式识别中手动提取特征会产生信息损失，针对脑电的时空特性，本文设计了具有时域卷积核、空间域卷积核的深层卷积神经网络和浅层卷积神经网络两种网络结构，将特征提取和状态分类合二为一，对正常态与疲劳态脑电数据进行分类，可视化了卷积神经网络的空间域卷积核。结果表明，浅层卷积神经网络平均分类正确率为98.868%，深层卷积神经网络平均分类正确率为98.217%，均高于传统分类方法，通过空间域卷积核的可视化，能够了解不同导联在网络中的参与程度，验证了该模型在脑疲劳检测任务中具有很高的有效性，同时为脑疲劳检测提供了新思路。      

Massively parallel classification of single-trial EEG signals using a min-max modular neural network

This paper presents a method for classifying single-trial electroencephalogram (EEG) signals using min-max modular neural networks implemented in a massively parallel way. The method has three main steps. First, a large-scale, complex EEG classification problem is simply divided into a reasonable number of two-class subproblems, as small as needed. Second, the two-class subproblems are simply learned by individual smaller network modules in parallel. Finally, all the individual trained network modules are integrated into a hierarchical, parallel, and modular classifier according to two module combination laws. To demonstrate the effectiveness of the method, we perform simulations on fifteen different four-class EEG classification tasks, each of which consists of 1491 training and 636 test data. These EEG classification tasks were created using a set of non-averaged, single-trial hippocampal EEG signals recorded from rats; the features of the EEG signals are extracted using wavelet transform techniques. The experimental results indicate that the proposed method has several attractive features. 1) The method is appreciably faster than the existing approach that is based on conventional multilayer perceptrons. 2) Complete learning of complex EEG classification problems can be easily realized, and better generalization performance can be achieved. 3) The method scales up to large-scale, complex EEG classification problems.     

Neural network approach to land cover mapping

A pattern classification method is proposed for remote sensing data using neural networks. First, the authors apply the error backpropagation (BP) algorithm to classify the remote sensing data. In this case, the classification performance depends on a training data set. In order to get stable and precise classification results, the training data set is selected based on geographical information and Kohonen's self-organizing feature map. Using the training data set and the error backpropagation algorithm, a layered neural network is trained such that the training patterns are classified with a specified accuracy. After training the neural network, some pixels are deleted from the original training data set if they are incorrectly classified and a new training data set is built up. Once training is complete, a testing data set is classified by using the trained neural network. The classification results of LANDSAT TM data show that this approach produces excellent results which are more realistic and noiseless compared with a conventional Bayesian method     

Detection of characteristic waves of sleep EEG by neural network analysis

In psychiatry, the sleep stage is one of the most important evidence for diagnosing mental disease. However, doctors require much labor and skill for diagnosis, so a quantitative and objective method is required for more accurate diagnosis since it depends on the doctor's experience. For this reason, an automatic diagnosis system must be developed. In this paper, we propose a new type of neural network (NN) model referred to as a sleep electroencephalogram (EEG) recognition neural network (SRNN) which enables us to detect several kinds of important characteristic waves in sleep EEG which are necessary for diagnosing sleep stages. Experimental results indicate that the proposed NN model was much more capable than other conventional methods for detecting characteristic waves.     

Multiclass support vector machines for EEG-signals classification.

In this paper, we proposed the multiclass support vector machine (SVM) with the error-correcting output codes for the multiclass electroencephalogram (EEG) signals classification problem. The probabilistic neural network (PNN) and multilayer perceptron neural network were also tested and benchmarked for their performance on the classification of the EEG signals. Decision making was performed in two stages: feature extraction by computing the wavelet coefficients and the Lyapunov exponents and classification using the classifiers trained on the extracted features. The purpose was to determine an optimum classification scheme for this problem and also to infer clues about the extracted features. Our research demonstrated that the wavelet coefficients and the Lyapunov exponents are the features which well represent the EEG signals and the multiclass SVM and PNN trained on these features achieved high classification accuracies.     

Optimizing the channel selection and classification accuracy in EEG-based BCI

Multichannel EEG is generally used in brain-computer interfaces (BCIs), whereby performing EEG channel selection 1) improves BCI performance by removing irrelevant or noisy channels and 2) enhances user convenience from the use of lesser channels. This paper proposes a novel sparse common spatial pattern (SCSP) algorithm for EEG channel selection. The proposed SCSP algorithm is formulated as an optimization problem to select the least number of channels within a constraint of classification accuracy. As such, the proposed approach can be customized to yield the best classification accuracy by removing the noisy and irrelevant channels, or retain the least number of channels without compromising the classification accuracy obtained by using all the channels. The proposed SCSP algorithm is evaluated using two motor imagery datasets, one with a moderate number of channels and another with a large number of channels. In both datasets, the proposed SCSP channel selection significantly reduced the number of channels, and outperformed existing channel selection methods based on Fisher criterion, mutual information, support vector machine, common spatial pattern, and regularized common spatial pattern in classification accuracy. The proposed SCSP algorithm also yielded an average improvement of 10% in classification accuracy compared to the use of three channels (C3, C4, and Cz).     

Learn++: an incremental learning algorithm for supervised neuralnetworks

We introduce Learn++, an algorithm for incremental training of neural network (NN) pattern classifiers. The proposed algorithm enables supervised NN paradigms, such as the multilayer perceptron (MLP), to accommodate new data, including examples that correspond to previously unseen classes. Furthermore, the algorithm does not require access to previously used data during subsequent incremental learning sessions, yet at the same time, it does not forget previously acquired knowledge. Learn++ utilizes ensemble of classifiers by generating multiple hypotheses using training data sampled according to carefully tailored distributions. The outputs of the resulting classifiers are combined using a weighted majority voting procedure. We present simulation results on several benchmark datasets as well as a real-world classification task. Initial results indicate that the proposed algorithm works rather well in practice. A theoretical upper bound on the error of the classifiers constructed by Learn++ is also provided     

面向癫痫脑电图信号识别的径向基最小最大概率分类树

脑电图(EEG)信号检测和识别是癫痫病的重要诊断手段。径向基函数神经网络具有出色的逼近能力和泛化性能,能直接识别出不同状态的脑电信号,但其透明性和可解释性差,忽视了不同类别数据间可分性的不同。对此,该文提出一种基于径向基函数神经网络和最小最大概率决策技术的分类树,采用一对一策略和排除法,更多考虑了类间可分性的不同。针对脑电信号识别的实验表明,所提方法结构清晰,分类能力强,可解释性更好。     

基于神经网络的光标上下移动EEG信号分类方法

脑电信号（EEG）是研究脑活动的一种重要的信息来源,基于脑电信号的人与计算机的通信已成为一种新的人机接口方式。在此主要通过时域回归方法对BCIⅡ竞赛数据进行EEG信号去噪预处理,运用6阶AR参数提取脑电特征作为神经网络的输入,最后用Matlab 7.0进行仿真,得到分类正确率为90%。实验表明,该方法可以达到很好的分类效果。     

GPS GDOP classification via improved neural network trainings and principal component analysis

Geometric dilution of precision (GDOP) is an engineering expression that denotes how well the constellation of global positioning system (GPS) satellites is organised geometrically. In the analysis of received signals, it is often essential to invert and transform the data matrices. This requires tremendous computational burden on the navigator’s processor. Since classification of GPS GDOP is a non-linear problem, neural networks (NNs) can be used as an acceptable solution. Since the back propagation (BP) does not have sufficient speed to train a feed-forward NN, in this paper several improved NN trainings, including Levenberg–Marquardt (LM), modified LM, and resilient BP (RBP), scaled conjugate gradient, one-step secant (OSS) and quasi-Newton methods are proposed to classify the GPS GDOP. In this study, in order to have uncorrelated and informative features of the GPS GDOP, principal component analysis (PCA) is used as a pre-processing step. The simulation results show that using the RBP and PCA leads to greater accuracy and lower calculation time comparing with other existing and proposed methods and it can improve the classification accuracy of GPS satellites to about 99.65%. Moreover, the modified LM is the fastest algorithm that requires only 10 iterations for training the NN and it can be used in online applications.     

基于相位延迟指数的脑功能网络及测谎研究

在脑认知科学领域,越来越多的研究开始专注于利用不同导联脑电信号之间的相互依赖关系来研究大脑整体认知功能.相位延迟指数可有效减少由容积导体引起的误差,该方法已被广泛应用.而基于图论的脑网络研究方法在测谎方面还少见报道.本文通过对30名（诚实和说谎）受试者的脑电信号进行网络拓扑分析,将网络参数作为判别指标,使用支持向量机对实验数据进行分类.研究发现,两组受试者的小世界指标表现出显著的统计学差异,且得到较高的测谎准确率,结果证明了利用相位延迟指数方法进行图论分析的测谎有效性.     

Extraction and classification of tempo stimuli from electroencephalography recordings using convolutional recurrent attention model

Electroencephalography (EEG) recordings taken during the perception of music tempo contain information that estimates the tempo of a music piece. If information about this tempo stimulus in EEG recordings can be extracted and classified, it can be effectively used to construct a music-based brain–computer interface. This study proposes a novel convolutional recurrent attention model (CRAM) to extract and classify features corresponding to tempo stimuli from EEG recordings of listeners who listened with concentration to the tempo of musics. The proposed CRAM is composed of six modules, namely, network inputs, two-dimensional convolutional bidirectional gated recurrent unit-based sample encoder, sample-level intuitive attention, segment encoder, segment-level intuitive attention, and softmax layer, to effectively model spatiotemporal features and improve the classification accuracy of tempo stimuli. To evaluate the proposed method's performance, we conducted experiments on two benchmark datasets. The proposed method achieves promising results, outperforming recent methods.     

一种模糊-证据kNN分类方法

 已有的以k-最近邻(k Nearest Neighbor,kNN)规则为核心的分类算法,如模糊kNN(Fuzzy kNN,FkNN)和证据kNN (Evidential kNN,EkNN)等,存在着两个问题:无法区别出样本特征的差异以及忽略了邻居距训练样本类中心距离的不同所带来的影响.为此,本文提出一种模糊-证据kNN算法.首先,利用特征的模糊熵值确定每个特征的权重,基于加权欧氏距离选取k个邻居;然后,利用邻居的信息熵区别对待邻居并结合FkNN在表示信息和EkNN在融合决策方面的优势,采取先模糊化再融合的方法确定待分类样本的类别.本文的方法在UCI标准数据集上进行了测试,结果表明该方法优于已有算法.     

The use of fuzzy integrals and bispectral analysis of the electroencephalogram to predict movement under anesthesia

The objective of this study was to design and evaluate a methodology for estimating the depth of anesthesia in a canine model that integrates electroencephalogram (EEG)-derived autoregressive (AR) parameters, hemodynamic parameters, and the alveolar anesthetic concentration. Using a parameters, and the alveolar anesthetic concentration. Using a parametric approach, two separate AR models of order ten were derived for the EEG, one from the third-order cumulant sequence and the other from the autocorrelation lags of the EEG. Since the anesthetic dose versus depth of anesthesia curve is highly nonlinear, a neural network (NN) was chosen as the basic estimator and a multiple NN approach was conceived which took hemodynamic parameters, EEG derived parameters, and anesthetic concentration as input feature vectors. Since the estimation of the depth of anesthesia involves cognitive as well as statistical uncertainties, a fuzzy integral was used to integrate the individual estimates of the various networks and to arrive at the final estimate of the depth of anesthesia. Data from 11 experiments were used to train the NN's which were then tested on nine other experiments. The fuzzy integral of the individual NN estimates (when tested on 43 feature vectors from seven of the nine test experiments) classified 40 (93%) of them correctly, offering a substantial improvement over the individual NN estimates.     

Classification of seismic signals by integrating ensembles ofneural networks

We examine a classification problem in which seismic waveforms of natural earthquakes are to be distinguished from waveforms of man-made explosions. We present an integrated classification machine (ICM), which is a hierarchy of artificial neural networks (ANNs) that are trained to classify the seismic waveforms. In order to maximize the gain of combining the multiple ANNs, we suggest construction of a redundant classification environment (RCE) that consists of several “experts” whose expertise depends on the different input representations to which they are exposed. In the proposed scheme, the experts are ensembles of ANN, trained on different bootstrap replicas. We use various network architectures, different time-frequency decompositions of the seismic waveforms, and various smoothing levels in order to achieve an RCE. A confidence measure for the ensemble's classification is defined based on the agreement (variance) within the ensembles, and an algorithm for a nonlinear integration of the ensembles using this measure is presented. An implementation on a data set of 380 seismic events is described, where the proposed ICM had classified correctly 92% of the testing signals. The comparison we made with classical methods indicates that combining a collection of ensembles of ANNs can be used to handle complex high dimensional classification problems     

在线脑机接口中脑电信号的特征提取与分类方法

在脑机接口研究中,针对运动想象脑电信号的特征抽取,提出了一种基于离散小波变换和AR模型的方法.利用Daubechies类小波函数对脑电信号进行3层分解,抽取小波变换系数的统计特征;利用Burg算法提取脑电信号6阶AR模型系数.将这两类特征进行组合后使用神经网络、支持向量机、马氏距离线性判别进行分类并比较分析.采用BCI...     

Utilization of multiple polarization data for aerospace targetidentification

Two identification approaches, the matching score (MS) method and the neural network (NN) method, and multiple polarization data are utilized to identify aerospace targets. A majority vote rule, a maximum isolation distance rule, and a combination of these two rules are proposed to determine a target class when four polarization combinations data are available. It is found that by combining the two decision rules, the recognition rates can be greatly improved. The effect of Gaussian noise on the recognition rates with the MS method and the NN method is also studied. If only the factor of Gaussian noise is considered, it is found that the MS method is more robust to Gaussian noise contamination than the NN method when the network is trained only by the uncontaminated range profiles     

A Neural Network Method for Traffic Dispersion

１ＩｎｔｒｏｄｕｃｔｉｏｎＴｒａｆｉｃｄｉｓｐｅｒｓｉｏｎ［１］，ｗｈｉｃｈｍｅａｎｓｔｈｅｔｒａｆｉｃｉｓｄｉｓｔｒｉｂｕｔｅｄｏｖｅｒｍｕｌｔｉｐｌｅｐａｔｈｓａｎｄｔｒａｎｓｍｉｔｅｄｉｎｐａｒａｌｅｌ，ｉｓｕｓｅｄｔｏｉｍｐｒｏｖｅｔｈｅｎｅ...     

混合卷积神经网络的高光谱图像分类方法

为了解决高光谱图像领域中,传统卷积神经网络因部分特征信息损失而影响最终地物分类精度的问题,采用一种基于2维和3维的混合卷积神经网络的高光谱图像分类方法,从空间增强、光谱-空间两方面分别进行了特征提取.首先从空间增强角度提出一种3维-2维卷积神经网络混合结构,得到增强后的空间信息;其次从光谱-空间角度利用3维卷积网络结构...     

Regularized common spatial pattern with aggregation for EEG classification in small-sample setting

Common spatial pattern (CSP) is a popular algorithm for classifying electroencephalogram (EEG) signals in the context of brain-computer interfaces (BCIs). This paper presents a regularization and aggregation technique for CSP in a small-sample setting (SSS). Conventional CSP is based on a sample-based covariance-matrix estimation. Hence, its performance in EEG classification deteriorates if the number of training samples is small. To address this concern, a regularized CSP (R-CSP) algorithm is proposed, where the covariance-matrix estimation is regularized by two parameters to lower the estimation variance while reducing the estimation bias. To tackle the problem of regularization parameter determination, R-CSP with aggregation (R-CSP-A) is further proposed, where a number of R-CSPs are aggregated to give an ensemble-based solution. The proposed algorithm is evaluated on data set IVa of BCI Competition III against four other competing algorithms. Experiments show that R-CSP-A significantly outperforms the other methods in average classification performance in three sets of experiments across various testing scenarios, with particular superiority in SSS.