Performance analysis of classical, model-based and eigenvector methods: Ophthalmic arterial disorders detection case

In this study, ophthalmic arterial Doppler signals recorded from 214 subjects were processed using classical, model-based, and eigenvector methods. The classical method (fast Fourier transform), two model-based methods (Burg autoregressive, least squares modified Yule-Walker autoregressive moving average methods), and three eigenvector methods (Pisarenko, multiple signal classification, and Minimum-Norm methods) were selected for performing spectral analysis of the ophthalmic arterial Doppler signals. Doppler power spectral density estimates of the ophthalmic arterial Doppler signals were obtained using these spectrum analysis techniques. The variations in the shape of the Doppler power spectra were examined in order to detect variabilities such as stenosis, ocular Behcet disease, and uveitis disease in the physical state of ophthalmic arterial Doppler signals. These power spectra were then used to compare the applied methods in terms of their frequency resolution and the effects in detecting stenosis, Behcet disease and uveitis disease in ophthalmic arteries.     

Selection of optimal AR spectral estimation method for internal carotid arterial Doppler signals using Cramer-Rao bound

In this study, Doppler signals recorded from internal carotid artery of 80 subjects were processed by PC-computer using autoregressive method and Doppler power spectra were obtained. The parameters of autoregressive method were estimated by different estimation methods such as Yule-Walker, covariance, modified covariance, Burg, least squares, and maximum likelihood estimation. Doppler spectra were then used to compare the applied estimation methods in terms of their frequency resolution and the effects in determination of stenosis in internal carotid arteries. The Cramer-Rao bounds were derived for the estimated autoregressive parameters of the internal carotid arterial Doppler signals and the performance evaluation of the estimation methods was performed using the Cramer-Rao bound values. Finally, the optimal autoregressive spectral estimation method for the internal carotid arterial Doppler signals was selected according to the computed Cramer-Rao bound values.     

Least squares support vector machine employing model-based methods coefficients for analysis of EEG signals

The aim of the study is classification of the electroencephalogram (EEG) signals by combination of the model-based methods and the least squares support vector machines (LS-SVMs). The LS-SVMs were implemented for classification of two types of EEG signals (set A – EEG signals recorded from healthy volunteers with eyes open and set E – EEG signals recorded from epilepsy patients during epileptic seizures). In order to extract the features representing the EEG signals, the spectral analysis of the EEG signals was performed by using the three model-based methods (Burg autoregressive – AR, moving average – MA, least squares modified Yule–Walker autoregressive moving average – ARMA methods). The present research demonstrated that the Burg AR coefficients are the features which well represent the EEG signals and the LS-SVM trained on these features achieved high classification accuracies.     

Analysis of the Doppler signals using largest Lyapunov exponent and correlation dimension in healthy and stenosed internal carotid artery patients

Nonlinear analysis techniques have been applied for many applications on the physiological systems. In this research, the largest Lyapunov exponent (LLE) and correlation dimension (CD) calculations were performed to evaluate the chaotic behaviour of blood flow obtained from the healthy and stenosed internal carotid artery (ICA) using noninvasive Doppler ultrasonography technique. The Doppler signals were taken from 30 healthy, 8 mild, 8 moderate and 8 serious degree of stenosis ICA subjects. The LLE calculation was performed by using Wolf algorithm. The Grassberger–Procaccia algorithm was used for CD analysis. The calculated LLE and CD values for stenosed ICA Doppler signals were found as significantly high compared to the values belonging to the healthy subjects. It is found that, the LLEs and CDs of stenosed ICA Doppler signals increase with the increasing of the degree of stenosis. The results show that the LLE and CD can be used for diagnosis of the ICA stenosis and to determine the degree of stenosis.     

EMGdi signal enhancement based on ICA decomposition and wavelet transform

Diaphragmatic electromyogram (EMGdi) signal plays an important role in the diagnosis and analysis of respiratory diseases. However, EMGdi recordings are often contaminated by electrocardiographic (ECG) interference, which posing serious obstacle to traditional denoising approaches due to overlapped spectra of these signals. In this paper, a novel method based on wavelet transform and independent component analysis (ICA) is proposed to remove the ECG interference from noisy EMGdi signals. With the proposed method, the original independent components of contaminated EMGdi signal were first obtained with ICA. Then the ECG components contained were removed by a specially designed wavelet domain filter. After that, the purified independent components were reconstructed back to the original signal space by ICA to obtain clean EMGdi signals. Experimental results achieved on practical clinical data show that the proposed approach is better than several traditional methods include wavelet transform (WT), ICA, digital filter and adaptive filter in ECG interference removing.     

Combining eigenvector methods and support vector machines for detecting variability of Doppler ultrasound signals

In this paper, the multiclass support vector machines (SVMs) with the error correcting output codes (ECOC) were presented for detecting variabilities of the multiclass Doppler ultrasound signals. The ophthalmic arterial (OA) Doppler signals were recorded from healthy subjects, subjects suffering from OA stenosis, subjects suffering from ocular Behcet disease. The internal carotid arterial (ICA) Doppler signals were recorded from healthy subjects, subjects suffering from ICA stenosis, subjects suffering from ICA occlusion. Methods of combining multiple classifiers with diverse features are viewed as a general problem in various application areas of pattern recognition. Because of the importance of making the right decision, better classification procedures for Doppler ultrasound signals are searched. Decision making was performed in two stages: feature extraction by eigenvector methods and classification using the SVMs trained on the extracted features. The research demonstrated that the multiclass SVMs trained on extracted features achieved high accuracy rates.     

Classification of transcranial Doppler signals using their chaotic invariant measures

In this study, chaos analysis was performed on the transcranial Doppler (TCD) signals recorded from the temporal region of the brain of 82 patients as well as of 24 healthy people. Two chaotic invariant measures, i.e. the maximum Lyapunov exponent and the correlation dimension, were calculated for the TCD signals after applying nonlinearity and stationarity tests to them. The sonograms obtained via Burg autoregressive (AR) method demonstrated that the chaotic invariant measures represented the unpredictability and complexity levels of the TCD signals. According to the multiple linear regression analysis, the chaotic invariant measures were found to be highly significant for the regression equation which fitted to the data. This result suggested that the chaotic invariant measures could be used for automatically differentiating various cerebrovascular conditions via an appropriate classifier. For comparison purposes, we investigated several different classification algorithms. The k-nearest neighbour algorithm outperformed all the other classifiers with a classification accuracy of 94.44% on the test data. We used the receiver operating characteristic (ROC) curves in order to assess the performance of the classifiers. The results suggested that the classification systems which use the chaotic invariant measures as input have potential in detecting the blood flow velocity changes due to various brain diseases.     

基于语音信号稀疏性的FDICA初始化和后处理方法

目前解决语音信号盲源分离(Blind source separation,BSS)的两大类方法分别为频域独立成分分析(Frequency domain independent component analysis,FDICA)和基于稀疏性的时频掩蔽(Time frequency masking,TF masking).为此将两类方法优点相结合,利用TF masking方法的结果,对FDICA做初始化,在加快FDICA收敛速度的同时也避免了次序不确定性问题.此外还提出了一种新的基于语音稀疏性FDICA的BSS后处理方法:基于局部最小比例控制(Local minimum ratio controlled,LMRC)谱减法,比常规的TF masking、维纳滤波等后处理方法,能够更有效地控制音乐噪声,提高分离性能.合成数据和实际采集数据的实验结果验证了所提方法的有效性.     

基于fMRI瞬时功率的独立成分分析

独立成分分析(independent component analysis,ICA)采用一种统计隐变量模型,假设信号是由各信源线性叠加构成.为了解决功能磁共振数据(functional magnetic resonance imaging,fMRI)中由于信源非线性叠加造成的ICA检测误差,提出了基于瞬时功率的ICA方法.首先,由电流能量形式将fMRI数据推广为fMRI能量信号;然后,由血氧水平依赖(blood oxygenation level dependent,BOLD)信号与T2*信号的关系,给出了两种反映BOLD能量变化的瞬时功率fMRI信号;最后,采用空间ICA分析fMRI瞬时功率信号,得到与各脑部活跃区域能量相关的独立成分.从理论和仿真试验两个方面阐明了新方法的合理性和优越性,同时应用于实际癫痫fMRI数据,经与传统ICA方法比较,该方法能够在静息态下鲁棒地检测脑部能量异常区域.     

Applications of ICA for the enhancement and classification of polarimetric SAR images

Independent components analysis (ICA) based methods for polarimetric synthetic aperture radar (SAR) image speckle reduction and ground object classification are studied. Several independent components can be extracted from polarimetric SAR images using ICA directly. The component with lowest speckle index is regarded as the scene after speckle reduction. The disadvantage of this method is that only one image is kept and most polarization information will be lost. In this paper, we use ICA‐sparse‐coding shrinkage (ICA‐SPS) based speckle reduction method, which is implemented on each individual image and can keep polarization information. It is carried out on the combined channels obtained by Pauli‐decomposition rather than original polarization channels in order to keep relative phase information among polarization channels and get better performance. After ICA‐SPS, the effect of speckle suppression on SAR image classification can be compared favourably with other methods by combining the channels into a false colour image. At last, a new ICA‐based classification method is presented. In this method, four independent components are separated by ICA from five polarization and combined channels. One of these independent components which includes little ground object information is regarded as speckle noise and therefore be discarded. The remaining three components can be treated as subordination coefficients of three kinds of targets. A classified image can be obtained based on the components. And by composing these three channels in RGB colour pattern, a false colour image can be constructed.     

基于独立成分的局部建模方法及其在近红外光谱分析中的应用研究

建立了一种基于独立成分分析的局部建模新方法,该方法首先将独立成分分析(ICA)用于近红外光谱的特征提取,然后,根据所提取的独立成分选择校正集中与预测样本相邻近的样本构成校正子集,建立局部偏最小二乘(PLS)回归模型并对预测样本进行预测。将所提出的方法应用于烟草样品中尼古丁含量的测定,所得结果优于常用的全局建模方法。     

Medical diagnosis of rheumatoid arthritis disease from right and left hand Ulnar artery Doppler signals using adaptive network based fuzzy inference system (ANFIS) and MUSIC method

Rheumatoid arthritis (RA) is a multi-systemic autoimmune disease that leads to substantial morbidity and mortality. In this paper, as spectral analysis methods of Multiple Signal Classification (MUSIC) method is used in order to extract the significant features from the right and left hand Ulnar artery Doppler signals for the diagnosis of RA disease. The MUSIC method has been used as subspace method. To extract features from Doppler signals obtained from the right and left hand Ulnar arterial the MUSIC method model degrees of 5, 10, 15, 20, and 25 were used. Then, an adaptive network based fuzzy inference system (ANFIS) was applied to features extracted from the right and left hand Ulnar artery Doppler signals for classifying RA disease. The methods are not new, but the study has a novelty in that the application area of these methods is new. In the hybrid model, the combination of MUSIC and ANFIS yielded classification accuracies of 95% (for a model degree of 20) using the right hand Ulnar artery and classification accuracies of 91.25% (for a model degree of 10) using left hand Ulnar artery Doppler signals in the diagnosis of RA disease. The proposed approach has potential to help with the early diagnosis of RA disease for the specialists who study this subject.     

Neural network analysis of ophthalmic arterial doppler signals with Uveitis disease

Doppler ultrasound is known as a reliable technique, which demonstrates the flow characteristics and resistance of ophthalmic arteries. In this study, ophthalmic arterial Doppler signals were obtained from 95 subjects—that 45 of them had suffered from Uveitis disease and the rest of them had been healthy subjects. Multilayer perceptron neural network (MLPNN) employing quick propagation training algorithm was used to detect the presence of Uveitis disease. Spectral analysis of ophthalmic arterial Doppler signals was performed by autoregressive moving average (ARMA) method for determining the MLPNN inputs. The MLPNN was trained with training set, cross validated with cross validation set and tested with testing set. All these data sets were obtained from ophthalmic arteries of healthy subjects and subjects suffering from Uveitis disease. Performance indicators and statistical measures were used for evaluating the MLPNN. The correct classification rate was 95.83% for healthy subjects and 91.30% for subjects suffering from Uveitis disease. Based on the accuracy of the MLPNN detections, it can be mentioned that the classification of ophthalmic arterial Doppler signals with Uveitis disease is feasible by the MLPNN employing quick propagation training algorithm.     

基于互累积量的有噪独立分量分析方法

针对传统独立分量分析(ICA)方法无噪假设的局限性,提出基于互累积量的有噪ICA方法。考虑含高斯噪声的瞬时混合模型,以观测信号的互累积量组成一系列对称矩阵,以对称矩阵的联合对角化程度为目标函数,采用粒子群优化算法对混合矩阵进行全局寻优。通过寻优得到混合矩阵,将有噪ICA转化为一维欠定ICA,基于奇异值分解法得到源信号的估计。仿真结果表明,与传统ICA方法相比,该方法对混合矩阵的估计精度较高,可以明显提高分离信号的信噪比。     

ICA based identification of dynamical systems generating synthetic and real world time series

Independent Component Analysis (ICA) is a recent and well known technique used to separate mixtures of signals. While in general the researchers put their attention on the type of signals and of mixing, we focus our attention on a quite general class of models which act as sources of the time series, the dynamical systems. In this paper we focus our attention on the general problem to understand the behaviour of ICA methods with respect to the time series deriving from a specific dynamical system, selecting large classes of them, and using ICA to make separation. This study gives some interesting results that are very useful both to highlight some properties related to dynamical systems and to clarify some general aspects of ICA, by using both synthetic and real data. From one hand we study the features of the linear (simple and coupled) and non-linear (single and coupled) dynamical systems, stochastic resonances, chaotic and real dynamical systems. We have to stress that we obtain information about the separation of these systems and substantially how from the entropy of the complete system we can obtain the entropies of the single dynamical systems (so that we also could obtain a more realistic analogic circuit). On the other hand these results show the high capability of the ICA method to recognize the dynamical systems independently from their complexity and in the case of stochastic series ICA perfectly recognizes the different dynamical systems also where the Fourier Transform is irresolute. We also note that in the case of real dynamical systems we showed that ICA permits to recognize the information connected to the sources and to associate to it a phenomenological dynamical system that reproduce it (i.e. Organ Pipe, Stromboli Volcano, Aerosol Index).     

An optimal approach for random signals classification

A method is proposed which solves the problem of the Bayes classification of ARMA (autoregressive moving average) signals when the models of classes and samples are not exactly known but only estimated from finite-length data sequences. Justified approximations and the hypothesis lead to decision rules including the variances of the estimations. The results obtained on a large set of simulated data show that this approach is superior to the best classical methods (cepstral distance or Kullback divergence), particularly in the common case where the hypothesis of those methods is not verified (short samples. small training sets. random classes)     

Model selection for convolutive ICA with an application to spatiotemporal analysis of EEG

We present a new algorithm for maximum likelihood convolutive independent component analysis (ICA) in which components are unmixed using stable autoregressive filters determined implicitly by estimating a convolutive model of the mixing process. By introducing a convolutive mixing model for the components, we show how the order of the filters in the model can be correctly detected using Bayesian model selection. We demonstrate a framework for deconvolving a subspace of independent components in electroencephalography (EEG). Initial results suggest that in some cases, convolutive mixing may be a more realistic model for EEG signals than the instantaneous ICA model.     

A fast algorithm for one-unit ICA-R

Independent component analysis (ICA) aims to recover a set of unknown mutually independent source signals from their observed mixtures without knowledge of the mixing coefficients. In some applications, it is preferable to extract only one desired source signal instead of all source signals, and this can be achieved by a one-unit ICA technique. ICA with reference (ICA-R) is a one-unit ICA algorithm capable of extracting an expected signal by using prior information. However, a drawback of ICA-R is that it is computationally expensive. In this paper, a fast one-unit ICA-R algorithm is derived. The reduction of the computational complexity for the ICA-R algorithm is achieved through (1) pre-whitening the observed signals; and (2) normalizing the weight vector. Computer simulations were performed on synthesized signals, a speech signal, and electrocardiograms (ECG). Results of these analyses demonstrate the efficiency and accuracy of the proposed algorithm.     

一种基于独立分量分析的模糊图像盲分离算法

利用独立分量分析（ICA）的不完整自然梯度算法对因混合而引起的多幅模糊灰度图像进行盲分离，并针对算法中的非线性函数与源信号概率分布密切相关，而源信号的分布却是未知的先验信息的问题，利用算法输出信号的峰度对非线性激活函数进行自适应选择，提出了一种改进的自适应不完整自然梯度算法，并将其应用于模糊图像的盲分离，分析了不同混合矩阵对本文算法恢复原始灰度图像的影响及算法性能。仿真结果证明了本文算法与经典的FastICA算法相比，计算耗时更少、性能指标明显优越。