EMG feature assessment for myoelectric pattern recognition and channel selection: A study with incomplete spinal cord injury

Myoelectric pattern recognition with a large number of electromyogram (EMG) channels provides an approach to assessing motor control information available from the recorded muscles. In order to develop a practical myoelectric control system, a feature dependent channel reduction method was developed in this study to determine a small number of EMG channels for myoelectric pattern recognition analysis. The method selects appropriate raw EMG features for classification of different movements, using the minimum Redundancy Maximum Relevance (mRMR) and the Markov random field (MRF) methods to rank a large number of EMG features, respectively. A k-nearest neighbor (KNN) classifier was used to evaluate the performance of the selected features in terms of classification accuracy. The method was tested using 57 channels’ surface EMG signals recorded from forearm and hand muscles of individuals with incomplete spinal cord injury (SCI). Our results demonstrate that appropriate selection of a small number of raw EMG features from different recording channels resulted in similar high classification accuracies as achieved by using all the EMG channels or features. Compared with the conventional sequential forward selection (SFS) method, the feature dependent method does not require repeated classifier implementation. It can effectively reduce redundant information not only cross different channels, but also cross different features in the same channel. Such hybrid feature-channel selection from a large number of EMG recording channels can reduce computational cost for implementation of a myoelectric pattern recognition based control system.     

GMDH-based feature ranking and selection for improved classification of medical data

Medical applications are often characterized by a large number of disease markers and a relatively small number of data records. We demonstrate that complete feature ranking followed by selection can lead to appreciable reductions in data dimensionality, with significant improvements in the implementation and performance of classifiers for medical diagnosis. We describe a novel approach for ranking all features according to their predictive quality using properties unique to learning algorithms based on the group method of data handling (GMDH). An abductive network training algorithm is repeatedly used to select groups of optimum predictors from the feature set at gradually increasing levels of model complexity specified by the user. Groups selected earlier are better predictors. The process is then repeated to rank features within individual groups. The resulting full feature ranking can be used to determine the optimum feature subset by starting at the top of the list and progressively including more features until the classification error rate on an out-of-sample evaluation set starts to increase due to overfitting. The approach is demonstrated on two medical diagnosis datasets (breast cancer and heart disease) and comparisons are made with other feature ranking and selection methods. Receiver operating characteristics (ROC) analysis is used to compare classifier performance. At default model complexity, dimensionality reduction of 22 and 54% could be achieved for the breast cancer and heart disease data, respectively, leading to improvements in the overall classification performance. For both datasets, considerable dimensionality reduction introduced no significant reduction in the area under the ROC curve. GMDH-based feature selection results have also proved effective with neural network classifiers.     

Comparison of feature evaluation criteria for speech recognition based on electromyography

In this paper, we present a performance comparison of 14 feature evaluation criteria and 4 classifiers for isolated Thai word classification based on electromyography signals (EMG) to find a near-optimal criterion and classifier. Ten subjects spoke 11 Thai number words in both audible and silent modes while the EMG signal from five positions of the facial and neck muscles were captured. After signal collection and preprocessing, 22 EMG features widely used in the EMG recognition field were computed and were then evaluated based on 14 evaluation criteria including both independent criteria (IC) and dependent criteria (DC) for feature evaluation and selection. Subsequently, the top nine features were selected for each criterion, and were used as inputs to classifiers. Four types of classifier were employed with 10-fold cross-validation to estimate classification performance. The results showed that features selected with a DC on a Fisher’s least square linear discriminant classifier (D_FLDA) used with a linear Bayes normal classifier (LBN) gave the best average accuracies, of 93.25 and 80.12% in the audible and the silent modes, respectively.     

Joint application of rough set-based feature reduction and Fuzzy LS-SVM classifier in motion classification

This paper presents an effective classification scheme consisting of the rough set theory (RST)-based feature selection and the fuzzy least squares support vector machine (LS-SVM) classifier for the surface electromyographic (sEMG)-based motion classification. The wavelet packet transform (WPT) is exploited to decompose the four-class motion EMG signals to the non-overlapped sub-bands and the energy characteristic of each sub-band is adopted to form the original feature set. In order to reduce the computation complexity, the RST is utilized to get the reduction feature set without compromising classification accuracy. In the feature reduction phase, cluster separation index (CSI) is introduced to evaluate the performance of the proposed algorithm. In the sequel, the Fuzzy LS-SVM is constructed for the multi-class classification task. The RST-based feature selection is independent of the classifier design. Consequently the classification performance will vary with different classifiers. We make the comparison between the proposed classification scheme and the commonly used classification scheme, such as the combination of the principal component analysis (PCA)-based feature selection and the neural network (NN) classifier. The results of comparative experiments show that the diverse motions can be identified with high accuracy by the proposed scheme. Compared with other feature extraction and selection algorithms and classifiers, superior performance of the proposed classification scheme illustrates the potential of the SVM techniques combined with WPT and RST in EMG motion classification.     

Feature-based classification of myoelectric signals using artificial neural networks

A pattern classification system, designed to separate myoelectric signal records based on contraction tasks, is described. The amplitude of the myoelectric signal during the first 200 ms following the onset of a contraction has a non-random structure that is specific to the task performed. This permits the application of advanced pattern recognition techniques to separate these signals. The pattern classification system described consists of a spectrographic preprocessor, a feature extraction stage and a classifier stage. The preprocessor creates a spectrogram by generating a series of power spectral densities over adjacent time segments of the input signal. The feature extraction stage reduces the dimensionality of the spectrogram by identifying features that correspond to subtle underlying structures in the input signal data. This is realised by a self-organising artificial neural network (ANN) that performs an advanced statistical analysis procedure known as exploratory projection pursuit. The extracted features are then classified by a supervised-learning ANN. An evaluation of the system, in terms of system performance and the complexity of the ANNs, is presented.     

Ant colony optimization-based feature selection method for surface electromyography signals classification

This paper presented a new ant colony optimization (ACO) feature selection method to classify hand motion surface electromyography (sEMG) signals. The multiple channels of sEMG recordings make the dimensionality of sEMG feature grow dramatically. It is known that the informative feature subset with small size is a precondition for the accurate and computationally efficient classification strategy. Therefore, this study proposed an ACO based feature selection scheme using the heuristic information measured by the minimum redundancy maximum relevance criterion (ACO-mRMR). The experiments were conducted on ten subjects with eight upper limb motions. Two feature sets, i.e., time domain features combined with autoregressive model coefficients (TDAR) and wavelet transform (WT) features, were extracted from the recorded sEMG signals. The average classification accuracies of using ACO reduced TDAR and WT features were 95.45±2.2% and 96.08±3.3%, respectively. The principal component analysis (PCA) was also conducted on the same data sets for comparison. The average classification accuracies of using PCA reduced TDAR and WT features were 91.51±4.9% and 89.87±4.4%, respectively. The results demonstrated that the proposed ACO-mRMR based feature selection method can achieve considerably high classification rates in sEMG motion classification task and be applicable to other biomedical signals pattern analysis.     

Classification of ankle joint movements based on surface electromyography signals for rehabilitation robot applications

Electromyography (EMG)-based control is the core of prostheses, orthoses, and other rehabilitation devices in recent research. Nonetheless, EMG is difficult to use as a control signal given the complex nature of the signal. To overcome this problem, the researchers employed a pattern recognition technique. EMG pattern recognition mainly involves four stages: signal detection, preprocessing feature extraction, dimensionality reduction, and classification. In particular, the success of any pattern recognition technique depends on the feature extraction stage. In this study, a modified time-domain features set and logarithmic transferred time-domain features (LTD) were evaluated and compared with other traditional time-domain features set (TTD). Three classifiers were employed to assess the two feature sets, namely linear discriminant analysis (LDA), k nearest neighborhood, and Naïve Bayes. Results indicated the superiority of the new time-domain feature set LTD, on conventional time-domain features TTD with the average classification accuracy of 97.23 %. In addition, the LDA classifier outperformed the other two classifiers considered in this study.     

A novel feature selection approach for biomedical data classification

This paper presents a novel feature selection approach to deal with issues of high dimensionality in biomedical data classification. Extensive research has been performed in the field of pattern recognition and machine learning. Dozens of feature selection methods have been developed in the literature, which can be classified into three main categories: filter, wrapper and hybrid approaches. Filter methods apply an independent test without involving any learning algorithm, while wrapper methods require a predetermined learning algorithm for feature subset evaluation. Filter and wrapper methods have their, respectively, drawbacks and are complementary to each other in that filter approaches have low computational cost with insufficient reliability in classification while wrapper methods tend to have superior classification accuracy but require great computational power. The approach proposed in this paper integrates filter and wrapper methods into a sequential search procedure with the aim to improve the classification performance of the features selected. The proposed approach is featured by (1) adding a pre-selection step to improve the effectiveness in searching the feature subsets with improved classification performances and (2) using Receiver Operating Characteristics (ROC) curves to characterize the performance of individual features and feature subsets in the classification. Compared with the conventional Sequential Forward Floating Search (SFFS), which has been considered as one of the best feature selection methods in the literature, experimental results demonstrate that (i) the proposed approach is able to select feature subsets with better classification performance than the SFFS method and (ii) the integrated feature pre-selection mechanism, by means of a new selection criterion and filter method, helps to solve the over-fitting problems and reduces the chances of getting a local optimal solution.     

A threshold fuzzy entropy based feature selection for medical database classification

Feature selection is one of the most common and critical tasks in database classification. It reduces the computational cost by removing insignificant features. Consequently, this makes the diagnosis process accurate and comprehensible. This paper presents the measurement of feature relevance based on fuzzy entropy, tested with a Radial Basis Function Network classifier for a medical database classification. Three feature selection strategies are devised to obtain the valuable subset of relevant features. Five benchmarked datasets, which are available in the UCI Machine Learning Repository, have been used in this work. The classification accuracy shows that the proposed method is capable of producing good results with fewer features than the original datasets.     

Supervised mutual-information based feature selection for motor unit action potential classification

A new supervised mutual information-based feature selection method is presented. Using real motor unit action potential (MUAP) data from 10 EMG signals, the performances of 32 time-sample feature sets, feature subsets selected using first- and second-order mutual information and features obtained using linear discriminant analysis (LDA) and principal component analysis (PCA) were evaluated using a minimum Euclidean distance (MED) classifier. The evaluation showed that by using only 20 first-order features or only 15 second-order features mean error rates and error rate variations equivalent to using all 32 samples or LDA or PCA could be obtained. The computational cost of first-order feature selection was considerably less than LDA, PCA and second-order feature selection. The performance of first-order features was further evaluated using a more robust classifier. Unlike the MED classifier, the robust classifier only assigned a candidate MUAP if the assignment was sufficiently certain. For the robust classifier the average error rates using 20 features were similar to using the full feature set, yet higher assignment rates were obtained. Results from both evaluations suggest that the sets of first-order features were an efficient representation of lower dimension, which provided high accuracy classification with reduced computational requirements.     

功能性磁共振成像在轻度认知障碍检测诊断的研究综述

现代社会中,阿尔茨海默病已经成为严重影响和限制个人日常生活甚至危及患者生命安全的一种疾病.轻度认知障碍作为阿尔茨海默病的前一个阶段,对其精确诊断有助于干预或降低患者转化为阿尔茨海默病的几率.目前,功能磁共振成像技术已经广泛应用于轻度认知障碍的检测诊断研究中.从特征提取、特征选择、数据降维和分类识别等方面,对fMRI在M...     

Automatic feature selection of motor imagery EEG signals using differential evolution and learning automata

Brain–computer interfacing (BCI) has been the most researched technology in neuroprosthesis in the last two decades. Feature extractors and classifiers play an important role in BCI research for the generation of suitable control signals to drive an assistive device. Due to the high dimensionality of feature vectors in practical BCI systems, implantation of efficient feature selection algorithms has been an integral area of research in the past decade. This article proposes an efficient feature selection technique, realized by means of an evolutionary algorithm, which attempts to overcome some of the shortcomings of several state-of-the-art approaches in this field. The outlined scheme produces a subset of salient features which improves the classification accuracy while maintaining a trade-off with the computational speed of the complete scheme. For this purpose, an efficient memetic algorithm has also been proposed for the optimization purpose. Extensive experimental validations have been conducted on two real-world datasets to establish the efficacy of our approach. We have compared our approach to existing algorithms and have established the superiority of our algorithm to the rest.     

Feature selection on movement imagery discrimination and attention detection

Noninvasive brain–computer interfaces (BCI) translate subject’s electroencephalogram (EEG) features into device commands. Large feature sets should be down-selected for efficient feature translation. This work proposes two different feature down-selection algorithms for BCI: (a) a sequential forward selection; and (b) an across-group variance. Power rar ratios (PRs) were extracted from the EEG data for movement imagery discrimination. Event-related potentials (ERPs) were employed in the discrimination of cue-evoked responses. While center-out arrows, commonly used in calibration sessions, cued the subjects in the first experiment (for both PR and ERP analyses), less stimulating arrows that were centered in the visual field were employed in the second experiment (for ERP analysis). The proposed algorithms outperformed other three popular feature selection algorithms in movement imagery discrimination. In the first experiment, both algorithms achieved classification errors as low as 12.5% reducing the feature set dimensionality by more than 90%. The classification accuracy of ERPs dropped in the second experiment since centered cues reduced the amplitude of cue-evoked ERPs. The two proposed algorithms effectively reduced feature dimensionality while increasing movement imagery discrimination and detected cue-evoked ERPs that reflect subject attention.     

Guilt-by-association feature selection: identifying biomarkers from proteomic profiles

In recent years, proteomic profiling by mass spectrometry has opened up a new realm of methods for identifying potential biomarkers. Mass spectrometry data, like other proteomic and genomic data, are challenging to analyze because of their high dimensionality and the availability of few samples. Hence, feature selection is extremely important because it directly provides a list of potential biomarkers by choosing a subset of effective features that separate diseased samples from healthy ones. The rule of thumb for feature selection is that features must be discriminant and independent for the best separation of the two groups. However, in general, existing feature selection algorithms only take into account the discrimination ability of features. In this paper, we present a novel method for feature selection, guilt-by-association feature selection (GBA-FS). The algorithm makes it possible to select features that are independent as well as discriminant. After measuring similarities between features, the algorithm groups together similar features using a clustering algorithm, and selects the best representative feature from each group. As a result, it produces a list of discriminant and independent features. The efficacy of GBA-FS was extensively tested on two real-world SELDI TOF data sets. The experimental results demonstrate that GBA-FS assists in selecting more independent features as compared to a common filter type feature selection method, the t test. The results also show that GBA-FS can be used to deconvolve multiply charged states of the same protein molecules. As GBA-FS successfully identifies feature groups with similar mass values, it can also be employed as an alternative to peak detection for preprocessing the mass spectrometry data.     

Evaluation of Karhunen-Loeve expansion for feature selection in computer-assisted classification of bioprosthetic heart-valve status

This paper analyses the performance of four different feature-selection approaches of the Karhunen-Loève expansion (KLE) method to select the most discriminant set of features for computer-assisted classification of bioprosthetic heart-valve status. First, an evaluation test reducing the number of initial features while maintaining the performance of the original classifier is developed. Secondly, the effectiveness of the classification in a simulated practical situation where a new sample has to be classified is estimated with a validation test. Results from both tests applied to a reference database show that the most efficient feature selection and classification (> or = 97% of correct classifications (CCs)) are performed by the Kittler and Young approach. For the clinical databases, this approach provides poor classification results for simulated 'new samples' (between 50 and 69% of CCs). For both the evaluation and the validation tests, only the Heydorn and Tou approach provides classification results comparable with those of the original classifier (a difference always < or = 7%). However, the degree of feature reduction is particularly variable. The study demonstrates that the KLE feature-selection approaches are highly population-dependent. It also shows that the validation method proposed is advantageous in clinical applications where the data collection is difficult to perform.     

基于脑肌电融合的混合脑机接口研究

动作模式识别是脑机接口技术的核心内容之一。针对目前脑机接口动作识别模式单一、识别率低等问题,基于混合脑机接口思想,提出一种脑电和肌电特征融合策略,可实现单侧肢体不同动作模式的有效分类,进而可用于脑机接口技术。同步采集9名健康受试者单侧手腕屈/伸两种动作模式下的脑电信号和表面肌电信号,分别提取脑电信号事件相关去同步化特征和表面肌电信号的积分肌电值特征,构建基于支持向量机和粒子群优化算法的脑肌电融合及运动模式识别模型,通过调整“特征融合系数”来实现动作模式最优分类,从而提高模式识别的准确率;进一步通过递降健康人的肌电信号幅值来模拟患者和运动疲劳状态下的肌电信号,验证所提出方法对动作模式识别的有效性。实验结果表明,基于脑肌电融合特征的动作模式识别率(98%)比单纯依靠脑电特征的识别率(73%)提高25%;在运动疲劳状态下,基于脑肌电融合特征的识别率稳定在80%以上,比单纯依靠肌电特征的识别率提高14%。可见,脑肌电融合策略能提高动作模式识别的准确性和鲁棒性,为混合脑机接口技术提供条件。     

A comparison of different feature extraction methods for diagnosis of valvular heart diseases using PCG signals

This article presents a novel method for diagnosis of valvular heart disease (VHD) based on phonocardiography (PCG) signals. Application of the pattern classification and feature selection and reduction methods in analysing normal and pathological heart sound was investigated. After signal preprocessing using independent component analysis (ICA), 32 features are extracted. Those include carefully selected linear and nonlinear time domain, wavelet and entropy features. By examining different feature selection and feature reduction methods such as principal component analysis (PCA), genetic algorithms (GA), genetic programming (GP) and generalized discriminant analysis (GDA), the four most informative features are extracted. Furthermore, support vector machines (SVM) and neural network classifiers are compared for diagnosis of pathological heart sounds. Three valvular heart diseases are considered: aortic stenosis (AS), mitral stenosis (MS) and mitral regurgitation (MR). An overall accuracy of 99.47% was achieved by proposed algorithm.     

Feature selection for optimized skin tumor recognition using genetic algorithms.

In this paper, a new approach to computer supported diagnosis of skin tumors in dermatology is presented. High resolution skin surface profiles are analyzed to recognize malignant melanomas and nevocytic nevi (moles), automatically. In the first step, several types of features are extracted by 2D image analysis methods characterizing the structure of skin surface profiles: texture features based on cooccurrence matrices, Fourier features and fractal features. Then, feature selection algorithms are applied to determine suitable feature subsets for the recognition process. Feature selection is described as an optimization problem and several approaches including heuristic strategies, greedy and genetic algorithms are compared. As quality measure for feature subsets, the classification rate of the nearest neighbor classifier computed with the leaving-one-out method is used. Genetic algorithms show the best results. Finally, neural networks with error back-propagation as learning paradigm are trained using the selected feature sets. Different network topologies, learning parameters and pruning algorithms are investigated to optimize the classification performance of the neural classifiers. With the optimized recognition system a classification performance of 97.7% is achieved.     

Classification of the myoelectric signal using time-frequency based representations

An accurate and computationally efficient means of classifying surface myoelectric signal patterns has been the subject of considerable research effort in recent years. Effective feature extraction is crucial to reliable classification and, in the quest to improve the accuracy of transient myoelectric signal pattern classification, an ensemble of time-frequency based representations are proposed. It is shown that feature sets based upon the short-time Fourier transform, the wavelet transform, and the wavelet packet transform provide an effective representation for classification, provided that they are subject to an appropriate form of dimensionality reduction.     

Conditioning and sampling issues of EMG signals in motion recognition of multifunctional myoelectric prostheses

Historically, the investigations of electromyography (EMG) pattern recognition-based classification of intentional movements for control of multifunctional prostheses have adopted the filter cut-off frequency and sampling rate that are commonly used in EMG research fields. In practical implementation of a multifunctional prosthesis control, it is desired to have a higher high-pass cut-off frequency to reduce more motion artifacts and to use a lower sampling rate to save the data processing time and memory of the prosthesis controller. However, it remains unclear whether a high high-pass cut-off frequency and a low-sampling rate still preserve sufficient neural control information for accurate classification of movements. In this study, we investigated the effects of high-pass cut-off frequency and sampling rate on accuracy in identifying 11 classes of arm and hand movements in both able-bodied subjects and arm amputees. Compared to a 5-Hz high-pass cut-off frequency, excluding the EMG components below 60 Hz decreased the average accuracy of 0.1% in classifying the 11 movements across able-bodied subjects and increased the average accuracy of 0.1 and 0.4% among the transradial (TR) and shoulder disarticulation (SD) amputees, respectively. Using a 500 Hz instead of a 1-kHz sampling rate, the average classification accuracy only dropped about 2.0% in arm amputees. The combination of sampling rate and high-pass cut-off frequency of 500 and 60 Hz only resulted in about 2.3% decrease in average accuracy for TR amputees and 0.4% decrease for SD amputees in comparison to the generally used values of 1 kHz and 5 Hz. These results suggest that the combination of sampling rate of 500 Hz and high-pass cut-off frequency of 60 Hz should be an optimal selection in EMG recordings for recognition of different arm movements without sacrificing too much of classification accuracy which can also remove most of motion artifacts and power-line interferences for improving the performance of myoelectric prosthesis control.