Classification of needle-EMG resting potentials by machine learning

Introduction: The diagnostic importance of audio signal characteristics in needle electromyography (EMG) is well established. Given the recent advent of audio-sound identification by artificial intelligence, we hypothesized that the extraction of characteristic resting EMG signals and application of machine learning algorithms could help classify various EMG discharges. Methods: Data files of 6 classes of resting EMG signals were divided into 2-s segments. Extraction of characteristic features (384 and 4,367 features each) was used to classify the 6 types of discharges using machine learning algorithms. Results: Across 841 audio files, the best overall accuracy of 90.4% was observed for the smaller feature set. Among the feature classes, mel-frequency cepstral coefficients (MFCC)-related features were useful in correct classification. Conclusions: We showed that needle EMG resting signals were satisfactorily classifiable by the combination of feature extraction and machine learning, and this can be applied to clinical settings. Muscle Nerve 59 :224–228, 2019     

Epileptic EEG classification based on extreme learning machine and nonlinear features

The automatic detection and classification of epileptic EEG are significant in the evaluation of patients with epilepsy. This paper presents a new EEG classification approach based on the extreme learning machine (ELM) and nonlinear dynamical features. The theory of nonlinear dynamics has been a powerful tool for understanding brain electrical activities. Nonlinear features extracted from EEG signals such as approximate entropy (ApEn), Hurst exponent and scaling exponent obtained with detrended fluctuation analysis (DFA) are employed to characterize interictal and ictal EEGs. The statistics indicate that the differences of those nonlinear features between interictal and ictal EEGs are statistically significant. The ELM algorithm is employed to train a single hidden layer feedforward neural network (SLFN) with EEG nonlinear features. The experiments demonstrate that compared with the backpropagation (BP) algorithm and support vector machine (SVM), the performance of the ELM is better in terms of training time and classification accuracy which achieves a satisfying recognition accuracy of 96.5% for interictal and ictal EEG signals.     

Automated classification of five seizure onset patterns from intracranial electroencephalogram signals

ObjectiveThe electroencephalographic (EEG) signals contain information about seizures and their onset location. There are several seizure onset patterns reported in the literature, and these patterns have clinical significance. In this work, we propose a system to automatically classify five seizure onset patterns from intracerebral EEG signals.MethodsThe EEG was segmented by clinicians indicating the start and end time of each seizure onset pattern, the channels involved at onset and the seizure onset pattern. Twelve features that represent the time domain characteristics and signal complexity were extracted from 663 seizures channels of 24 patients. The features were used for classification of the patterns with support vector machine - Error-Correcting Output Codes (SVM-ECOC). Three patient groups with a similar number of seizure segments were created, and one group was used for testing and the rest for training. This test was repeated by rotating the testing and training data.ResultsThe feature space formed by both time domain and multiscale sample entropy features perform well in classification of the data. An overall accuracy of 80.7% was obtained with these features and a linear kernel of SVM-ECOC.ConclusionsThe seizure onset patterns consist of varied time and complexity characteristics. It is possible to automatically classify various seizure onset patterns very similarly to visual classification.SignificanceThe proposed system could aid the medical team in assessing intracerebral EEG by providing an objective classification of seizure onset patterns.     

Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering

ObjectiveAccurate and reliable detection of tremor onset in Parkinson’s disease (PD) is critical to the success of adaptive deep brain stimulation (aDBS) therapy. Here, we investigated the potential use of feature engineering and machine learning methods for more accurate detection of rest tremor in PD.MethodsWe analyzed the local field potential (LFP) recordings from the subthalamic nucleus region in 12 patients with PD (16 recordings). To explore the optimal biomarkers and the best performing classifier, the performance of state-of-the-art machine learning (ML) algorithms and various features of the subthalamic LFPs were compared. We further used a Kalman filtering technique in feature domain to reduce the false positive rate.ResultsThe Hjorth complexity showed a higher correlation with tremor, compared to other features in our study. In addition, by optimal selection of a maximum of five features with a sequential feature selection method and using the gradient boosted decision trees as the classifier, the system could achieve an average F1 score of up to 88.7% and a detection lead of 0.52 s. The use of Kalman filtering in feature space significantly improved the specificity by 17.0% (p = 0.002), thereby potentially reducing the unnecessary power dissipation of the conventional DBS system.ConclusionThe use of relevant features combined with Kalman filtering and machine learning improves the accuracy of tremor detection during rest.SignificanceThe proposed method offers a potential solution for efficient on-demand stimulation for PD tremor.     

EEG-based motor imagery analysis using weighted wavelet transform features

In this study, an electroencephalogram (EEG) analysis system for single-trial classification of motor imagery (MI) data is proposed. Feature extraction in brain-computer interface (BCI) work is an important task that significantly affects the success of brain signal classification. The continuous wavelet transform (CWT) is applied together with Student's two-sample t-statistics for 2D time-scale feature extraction, where features are extracted from EEG signals recorded from subjects performing left and right MI. First, we utilize the CWT to construct a 2D time-scale feature, which yields a highly redundant representation of EEG signals in the time-frequency domain, from which we can obtain precise localization of event-related brain desynchronization and synchronization (ERD and ERS) components. We then weight the 2D time-scale feature with Student's two-sample t-statistics, representing a time-scale plot of discriminant information between left and right MI. These important characteristics, including precise localization and significant discriminative ability, substantially enhance the classification of mental tasks. Finally, a correlation coefficient is used to classify the MI data. Due to its simplicity, it will enable the performance of our proposed method to be clearly demonstrated. Compared to a conventional 2D time-frequency feature and three well-known time-frequency approaches, the experimental results show that the proposed method provides reliable 2D time-scale features for BCI classification.     

Detection and classification of subject-generated artifacts in EEG signals using autoregressive models

We examine the problem of accurate detection and classification of artifacts in continuous EEG recordings. Manual identification of artifacts, by means of an expert or panel of experts, can be tedious, time-consuming and infeasible for large datasets. We use autoregressive (AR) models for feature extraction and characterization of EEG signals containing several kinds of subject-generated artifacts. AR model parameters are scale-invariant features that can be used to develop models of artifacts across a population. We use a support vector machine (SVM) classifier to discriminate among artifact conditions using the AR model parameters as features. Results indicate reliable classification among several different artifact conditions across subjects (approximately 94%). These results suggest that AR modeling can be a useful tool for discriminating among artifact signals both within and across individuals.     

Analysis of electrooculography signals for the detection of Myasthenia Gravis

ObjectiveA precursor to more severe forms of Myasthenia Gravis (MG) is ocular MG (OMG) in which the MG symptoms are localized to the eyes. Current MG diagnostic methods are often invasive, painful, and not always specific. The objective of the proposed work was to extract quantifiable features from electrooculography (EOG) signals recorded around the eyes and develop an alternative non-invasive screening method for detecting MG.MethodsEOG signals acquired from MG and Control subjects were analyzed for eye movement characteristics and quantified using time and wavelet domain signal processing techniques. The ability of the proposed approaches to classify MG vs. control subjects was evaluated using a linear discriminant analysis (LDA) based classifier.ResultsThe range of overall classification accuracies achieved by the proposed time and wavelet domain approaches for different groupings were between 82.1–83.3% (Rise Rate feature: P < 0.01, AUC ≥ 0.87) and 82.1–87.2% (Mean Scale Band Energy feature: P < 0.01, AUC ≥ 0.89), respectively.ConclusionOur results demonstrate that an EOG-based signal analysis is a potentially viable non-invasive alternative for MG screening.SignificanceThe proposed approach could lead to early detection of MG and thereby improve clinical outcomes in this population.     

Dynamic functional connectivity in temporal lobe epilepsy: a graph theoretical and machine learning approach

Purpose

Functional magnetic resonance imaging (fMRI) in resting state can be used to evaluate the functional organization of the human brain in the absence of any task or stimulus. The functional connectivity (FC) has non-stationary nature and consented to be varying over time. By considering the dynamic characteristics of the FC and using graph theoretical analysis and a machine learning approach, we aim to identify the laterality in cases of temporal lobe epilepsy (TLE).

Methods

Six global graph measures are extracted from static and dynamic functional connectivity matrices using fMRI data of 35 unilateral TLE subjects. Alterations in the time trend of the graph measures are quantified. The random forest (RF) method is used for the determination of feature importance and selection of dynamic graph features including mean, variance, skewness, kurtosis, and Shannon entropy. The selected features are used in the support vector machine (SVM) classifier to identify the left and right epileptogenic sides in patients with TLE.

Results

Our results for the performance of SVM demonstrate that the utility of dynamic features improves the classification outcome in terms of accuracy (88.5% for dynamic features compared with 82% for static features). Selecting the best dynamic features also elevates the accuracy to 91.5%.

Conclusion

Accounting for the non-stationary characteristics of functional connectivity, dynamic connectivity analysis of graph measures along with machine learning approach can identify the temporal trend of some specific network features. These network features may be used as potential imaging markers in determining the epileptogenic hemisphere in patients with TLE.

     

An Insight into Extreme Learning Machines: Random Neurons,Random Features and Kernels

Extreme learning machines (ELMs) basically give answers to two fundamental learning problems: (1) Can fundamentals of learning (i.e., feature learning, clustering, regression and classification) be made without tuning hidden neurons (including biological neurons) even when the output shapes and function modeling of these neurons are unknown? (2) Does there exist unified framework for feedforward neural networks and feature space methods? ELMs that have built some tangible links between machine learning techniques and biological learning mechanisms have recently attracted increasing attention of researchers in widespread research areas. This paper provides an insight into ELMs in three aspects, viz: random neurons, random features and kernels. This paper also shows that in theory ELMs (with the same kernels) tend to outperform support vector machine and its variants in both regression and classification applications with much easier implementation.     

Epilepsy classification using optimized artificial neural network

ABSTRACT

Objectives: An Electroencephalogram (EEG) is the result of co-operative actions performed by brain cells. In other words, it can be defined as the time course of extracellular field potentials that are generated due to the synchronous action of cells. It is widely used for the analysis and diagnosis of several conditions. But this clinical data use to be multi-dimensional, context-dependent, complex, and it causes a concoction of various computing related research challenges. The objective of this study was to develop a computer-aided diagnosis system for epilepsy detection using EEG signals to ease the diagnosis process.

Materials: In this study, EEG datasets for epilepsy disease detection were taken from a public domain (Bonn University, Germany). These EEG recordings contain 100 single-channel EEG signals with maximum duration of 23.6 seconds. This data set was recorded intra-cranially and extra-cranially with the help of a 128-channel amplifier system using a common reference point.

Results: For a unique set of EEG signal features, the Optimized Artificial Neural Network model for classification and validation was developed with optimum neurons in the hidden layer. Results were tested on the basis of accuracy, sensitivity, precision, and specificity for all classes. The proposed Particle Swarm Optimized Artificial Neural Network provided 99.3% accuracy for EEG signal classification.

Discussion: Our results indicate that artificial neural network has efficiency to provide higher accuracy for epilepsy detection if the statistical features are extracted carefully. It is also possible to improve results for real time diagnosis by using optimization technique for error reduction.

Abbreviations: EEG: Electroencephalogram CAD: Computer-Aided Diagnosis ANN: Artificial Neural Network PSO: Particle Swarm Optimization FIR: Finite Impulse Response IIR: Infinite Impulse Response MSE: Mean Square Error.     

A Voting Optimized Strategy Based on ELM for Improving Classification of Motor Imagery BCI Data

This paper presents an approach to classifying electroencephalogram (EEG) signals for brain–computer interfaces (BCI). To eliminate redundancy in high-dimensional EEG signals and reduce the coupling among different classes of EEG signals, we use principle component analysis and linear discriminant analysis to extract features that represent the raw signals. Next, we introduce the voting-based extreme learning machine to classify the features. Experiments performed on real-world data from the 2003 BCI competition indicate that our classification method outperforms state-of-the-art methods in speed and accuracy.     

Applications of advanced signal processing and machine learning in the neonatal hypoxic-ischemic electroencephalography

Perinatal hypoxic-ischemic-encephalopathy significantly contributes to neonatal death and life-long disability such as cerebral palsy. Advances in signal processing and machine learning have provided the research community with an opportunity to develop automated real-time identification techniques to detect the signs of hypoxic-ischemic-encephalopathy in larger electroencephalography/amplitude-integrated electroencephalography data sets more easily. This review details the recent achievements, performed by a number of prominent research groups across the world, in the automatic identification and classification of hypoxic-ischemic epileptiform neonatal seizures using advanced signal processing and machine learning techniques. This review also addresses the clinical challenges that current automated techniques face in order to be fully utilized by clinicians, and highlights the importance of upgrading the current clinical bedside sampling frequencies to higher sampling rates in order to provide better hypoxic-ischemic biomarker detection frameworks. Additionally, the article highlights that current clinical automated epileptiform detection strategies for human neonates have been only concerned with seizure detection after the therapeutic latent phase of injury. Whereas recent animal studies have demonstrated that the latent phase of opportunity is critically important for early diagnosis of hypoxic-ischemic-encephalopathy electroencephalography biomarkers and although difficult, detection strategies could utilize biomarkers in the latent phase to also predict the onset of future seizures.     

Removal of Electrooculogram Artifacts from Electroencephalogram Using Canonical Correlation Analysis with Ensemble Empirical Mode Decomposition

Electrooculogram (EOG) is one of the major artifacts in the design of electroencephalogram (EEG)-based brain computer interfaces (BCIs). That removing EOG artifacts automatically while retaining more neural data will benefit for further feature extraction and classification. In order to remove EOG artifacts automatically as well as reserve more useful information from raw EEG, this paper proposes a novel blind source separation method called CCA-EEMD (canonical correlation analysis, ensemble empirical mode decomposition). Technically, the major steps of CCA-EEMD are as follows: Firstly, the multiple-channel original EEG signals are separated into several uncorrelated components using CCA. Then, the EOG component can be identified automatically by its kurtosis value. Next, the identified EOG component is decomposed into several intrinsic mode functions (IMFs) by EEMD. The IMFs uncorrelated to the EOG component are recognized and retained, and a new component will be constructed by the retained IMFs. Finally, the clean EEG signals are reconstructed. Keep in mind that the novelty of this paper is that the identified EOG component is not removed directly but used to extract neural EEG data, which would keep more effective information. Our tests with the data of seven subjects demonstrate that the proposed method has distinct advantages over other two commonly used methods in terms of average root mean square error [37.71 ± 0.14 (CCA-EEMD), 44.72 ± 0.13 (CCA), 49.59 ± 0.16 (ICA)], signal-to-noise ratio [3.59 ± 0.24 (CCA-EEMD), ?6.53 ± 0.18(CCA), ?8.43 ± 0.26 (ICA)], and classification accuracy [0.88 ± 0.002 (CCA-EEMD), 0.79 ± 0.001 (CCA), 0.73 ± 0.002 (ICA)]. The proposed method can not only remove EOG artifacts automatically but also keep the integrity of EEG data to the maximum extent.     

Anatomical Pattern Analysis for Decoding Visual Stimuli in Human Brains

A universal unanswered question in neuroscience and machine learning is whether computers can decode the patterns of the human brain. Multi-Voxel Pattern Analysis (MVPA) is a critical tool for addressing this question. However, there are two challenges in the previous MVPA methods, which include decreasing sparsity and noise in the extracted features and increasing the performance of prediction. In overcoming mentioned challenges, this paper proposes Anatomical Pattern Analysis (APA) for decoding visual stimuli in the human brain. This framework develops a novel anatomical feature extraction method and a new imbalance AdaBoost algorithm for binary classification. Further, it utilizes an Error-Correcting Output Codes (ECOC) method for multiclass prediction. APA can automatically detect active regions for each category of the visual stimuli. Moreover, it enables us to combine homogeneous datasets for applying advanced classification. Experimental studies on four visual categories (words, consonants, objects, and scrambled photos) demonstrate that the proposed approach achieves superior performance to state-of-the-art methods.     

A Novel Approach for Detecting Anomalous Energy Consumption Based on Micro-Moments and Deep Neural Networks

Nowadays, analyzing, detecting, and visualizing abnormal power consumption behavior of householders are among the principal challenges in identifying ways to reduce power consumption. This paper introduces a new solution to detect energy consumption anomalies based on extracting micro-moment features using a rule-based model. The latter is used to draw out load characteristics using daily intent-driven moments of user consumption actions. Besides micro-moment features extraction, we also experiment with a deep neural network architecture for efficient abnormality detection and classification. In the following, a novel anomaly visualization technique is introduced that is based on a scatter representation of the micro-moment classes, and hence providing consumers an easy solution to understand their abnormal behavior. Moreover, in order to validate the proposed system, a new energy consumption dataset at appliance level is also designed through a measurement campaign carried out at Qatar University Energy Lab, namely, Qatar University dataset. Experimental results on simulated and real datasets collected at two regions, which have extremely different climate conditions, confirm that the proposed deep micro-moment architecture outperforms other machine learning algorithms and can effectively detect anomalous patterns. For example, 99.58% accuracy and 97.85% F1 score have been achieved under Qatar University dataset. These promising results establish the efficacy of the proposed deep micro-moment solution for detecting abnormal energy consumption, promoting energy efficiency behaviors, and reducing wasted energy.

     

Feature Selection Based on Modified Bio-inspired Atomic Orbital Search Using Arithmetic Optimization and Opposite-Based Learning

Feature selection (FS) has the largest influence on the performance of machine learning methods. FS can remove the irrelevant and redundancy features from the data while preserving the same quality of increasing it. However, the traditional FS methods are time-consuming and can be stuck in local optima. So, the metaheuristic (MH) techniques are used to avoid these limitations since they have several operators that explore and exploit the search domain better than traditional methods. Besides these behaviors of MH, we present an improved atomic orbital search (IAOS) algorithm using a global search strategy that uses the operators of arithmetic optimization algorithm (AOA), which has proven a good exploration ability to provide a promising candidate solution. The opposite-based learning (OBL) is applied to enhance the initial population, which leads to enhancing the convergence rate towards the optimal solution. In addition, a dynamic photon rate is used to enhance the balance between exploration and exploitation. Finally, the sequential backward selection (SBS) is used as a local search strategy to improve the best solution, and this leads to obtaining a set of relevant features that increase the classification accuracy. To evaluate the performance of the presented IAOS-SBS as an FS method, a set of twenty UCI datasets is used; also, it is compared with other well-known FS methods. The results show the superiority of IAOS-SBS among the performance measures. Finally, it is concluded that IAOS-SBS can select fewer features with achieving high classification accuracy for most of the datasets utilized in the study. This indicates the use of OBL and SBS leads to enhancing the original AOS.

     

基于遗传算法与支持向量机的基因微阵列分析

背景：微阵列数据的特点是样本含量小，而变量数(基因)多达上万个。此时，传统的统计方法往往因为高维而失效了。遗传算法和支持向量机是近年来发展迅速的机器学习算法，具有很好的分类效果与降维优势。 目的：提出将遗传算法与支持向量机结合起来对样本进行分类，并与直接采用支持向量机、筛选差异表达基因后采用支持向量机的结果进行比较。 方法：采用Bioconductor提供的数据集golub,它是白血病微阵列芯片实验所得的基因表达数据集，对全部基因采用支持向量机进行分类。采用SAM软件对芯片数据的显著性分析确定不同的差异表达基因并估计错误发现率FDR，以筛选出的76个差异表达基因作为特征基因子集，再采用支持向量机进行分类。将筛选出的76个差异表达基因作为初始的特征基因集合，采用遗传算法－支持向量机再次进行特征基因选择，提高分类准确度，并与全部基因直接采用支持向量机、筛选差异表达基因后采用支持向量机的结果进行比较。同时也对特征基因在代谢通路上的分布和功能作了一定的研究。 结果与结论：通过遗传算法降维可以提高支持向量机的分类准确率，特别是剔除了数据中的大量无关基因和噪声，使得经过特征选择后分类准确率提高。结果显示遗传算法与支持向量机结合方法对分类更加有效。此外，通路分析结果显示特征基因的主要功能体现在信号传导和氨基酸代谢上。     

Noise-robust acoustic signature recognition using nonlinear Hebbian learning

We propose using a new biologically inspired approach, nonlinear Hebbian learning (NHL), to implement acoustic signal recognition in noisy environments. The proposed learning processes both spectral and temporal features of input acoustic data. The spectral analysis is realized by using auditory gammatone filterbanks. The temporal dynamics is addressed by analyzing gammatone-filtered feature vectors over multiple temporal frames, which is called a spectro-temporal representation (STR). Given STR features, the exact acoustic signatures of signals of interest and the mixing property between signals of interest and noises are generally unknown. The nonlinear Hebbian learning is then employed to extract representative independent features from STRs, and to reduce their dimensionality. The extracted independent features of signals of interest are called signatures. In the meantime of learning, the synaptic weight vectors between input and output neurons are adaptively updated. These weight vectors project data into a feature subspace, in which signals of interest are selected, while noises are attenuated. Compared with linear Hebbian learning (LHL) which explores the second-order moment of data, the applied NHL involves the higher-order statistics of data. Therefore, NHL can capture representative features that are more statistically independent than LHL can. Besides, the nonlinear activation function of NHL can be chosen to refer to the implicit distribution of many acoustic sounds, and thus making the learning optimized in an aspect of mutual information.Simulation results show that the whole proposed system can more accurately recognize signals of interest than other conventional methods in severely noisy circumstances. One applicable project is detecting moving vehicles. Noise-contaminated vehicle sound is recognized while other non-vehicle sounds are rejected. When vehicle is contaminated by human vowel, bird chirp, or additive white Gaussian noise (AWGN) at SNR=0 dB, the proposed system dramatically decreases the error rate over normally used acoustic feature extraction method, mel-frequency cepstral computation (MFCC), by 26%, 36.3%, and 60.3%, respectively; and, over LHL by 20%, 2.3%, and 15.3%, respectively. Another applicable project is vehicle type identification. The proposed system achieves better performance than LHL, e.g., 40% improvement when gasoline heavy wheeled car is contaminated by AWGN at SNR=5 dB. More importantly, the proposed system is implemented in real-time field testing for months. The purpose is to detect vehicle with any make or model moving on the street with speed 10–35 mph. The missing rate is 1–2%, when vehicle is contaminated by any surrounding noises (human conversation, animal sound, airplane, wind, etc.) at SNR=0–20 dB. The false alarm rate is around 1%.To summarize, this study not only provides an efficient approach to extract representative independent features from high-dimensional data, but also offers robustness against severe noises.     

Three-Way Image Classification with Evidential Deep Convolutional Neural Networks

The farfetched certain classification of uncertain data suffers serious risks. Three-Way Decision (3WD) theory is utilized to implement uncertain data classification methods. Three-way uncertain data classification methods facilitate reducing decision risk and involving human–machine coordination through finding out uncertain cases for abstaining identification. Due to the limitation of traditional classifiers in feature learning, most existing three-way uncertain data classification methods are not good at handling the unstructural data of digital images. This shortage hinders the applications of three-way uncertain data classification in image-based decision support systems, such as the medical decision support systems based on radiographs. In this paper, we adopt deep convolutional neural networks (DCNNs) for feature learning and Dempster–Shafer (D-S) evidence theory as uncertainty measure to implement a three-way method for image classification. We utilize evidence theory to measure the uncertainty of the predictions produced by DCNNs and construct a novel evidential deep convolutional neural network (EviDCNN). Based on this, we propose a Three-Way Classification method with EviDCNN (EviDCNN-3WC). The experiments on massive medical image data sets validate that the proposed three-way classification method with EviDCNN is effective to identify uncertain images and reduce the risk in image classification. The superiorities of the proposed method facilitate its applications in image-based medical decision support systems.

     

FE-ELM: A New Friend Recommendation Model with Extreme Learning Machine

Friend recommendation is one of the most popular services in location-based social network (LBSN) platforms, which recommends interested or familiar people to users. Except for the original social property and textual property in social networks, LBSN specially owns the spatial-temporal property. However, none of the existing methods fully utilized all the three properties (i.e., just one or two), which may lead to the low recommendation accuracy. Moreover, these existing methods are usually inefficient. In this paper, we propose a new friend recommendation model to solve the above shortcomings of the existing methods, called feature extraction-extreme learning machine (FE-ELM), where friend recommendation is regarded as a binary classification problem. Classification is an important task in cognitive computation community. First, we use new strategies in our FE-ELM model to extract the spatial-temporal feature, social feature, and textual feature. These features make full use of all above properties of LBSN and ensure the recommendation accuracy. Second, our FE-ELM model also takes advantage of the extreme learning machine (ELM) classifier. ELM has fast learning speed and ensures the recommendation efficiency. Extensive experiments verify the accuracy and efficiency of FE-ELM model.