Multiscale Two-Directional Two-Dimensional Principal Component Analysis and Its Application to High-Dimensional Biomedical Signal Classification

Abstract

Goal: Time-frequency analysis incorporating the wavelet transform followed by principal component analysis (WT-PCA) has been a powerful approach for the analysis of biomedical signals such as electromyography (EMG), electroencephalography (EEG), electrocardiography (ECG), and Doppler ultrasound. Time-frequency coefficients at various scales were usually transformed into a one-dimensional array using only a single or a few signal channels. The steady improvement of biomedical recording techniques has increasingly permitted the registration of a high number of channels. However, WT-PCA is not applicable to high-dimensional recordings due to the curse of dimensionality and small sample size problem. In this study, we present a multi-scale two-directional two-dimensional principal component analysis (MS2D 2 PCA) method for the efficient and effective extraction of essential feature information from high-dimensional signals. Multi-scale matrices constructed in the first step incorporate the spatial correlation and physiological characteristics of sub-band signals among channels. In the second step, the two-directional two-dimensional principal component analysis operates on the multi-scale matrices to reduce the dimension, rather than vectors in conventional PCA. Results are presented from an experiment to classify 20 hand movements using 89-channel EMG signals recorded in stroke survivors, which illustrates the efficiency and effectiveness of the proposed method for high-dimensional biomedical signal analysis.

Real-Time Embedded EMG Signal Analysis for Wrist-Hand Pose Identification

“…The final step in the EMG processing sequence is inference of movements from the reduceddimensionality feature vectors. Various classifiers have been analysed in this context, including Support Vector Machines (SVMs) [9], [10], [18], [21], [22], bayesian classifiers [14], gaussian mixture models [14], hidden markov models [14], fuzzy classifiers [14] and Multilayer Perceptrons (MLPs) [8], [16], [20], [23].…”

Exploration of Feature Extraction Methods and Dimension for sEMG Signal Classification

“…Therefore, ten time-domain features were selected for feature calculation. They were the mean absolute value (MAV) [15], root mean square (RMS) [16], Wilson amplitude (WAMP) [15], zero-crossing (ZC) [17], waveform length (WL) [18], the mean absolute value slope (MAVS) [30], slope sign change (SSC) [31], variance (VAR) [32], autoregressive (AR) coefficient of sEMG signal [27], and the AR coefficients from the first difference (FDAR) of the sEMG signal.…”

“…Such that the high-dimensional features can be reduced using a supervised feature reduction algorithm. Xie et al [27] have proposed a multi-scale bidirectional two-dimensional PCA method for high-dimensional biomedical signal analysis, which realized the dimensionality reduction of high-dimensional signals and proved to be highly efficient. Wang et al [28] have applied LDA classify the sEMG signals from 2 electrodes and recognize the grasping force.…”

Myoelectric signal classification based on S transform and two-directional two-dimensional principal component analysis

“…It is thus feasible to apply image processing techniques to indicate time-frequency matrix (TFM) characteristics. In a recent study, we have demonstrated the superiority of 2D 2 PCA over PCA to extract discriminant features form wavelet coefficients for high-density electrode array MES recognition (Xie et al, 2016). Despite the success of this recent 2D 2 PCA-based MES study, the involved wavelet transform exhibits some disadvantages, such as its complicated computation, sensitivity to noise level and the dependency of its accuracy on the chosen basis wavelet (Nguyen and Liao, 2009).…”

“…To evaluate the performance of the proposed method, results are presented on the recognition of eight hand motions from 4-channel MESs recorded in both health subjects and amputees, aiming for the prosthetic hand, robot, and human man interface controlling. The results obtained using ST2D 2 PCA are compared with WTPCA (Huang et al, 2012), WT2D 2 PCA (Xie et al 2016), as well as the hybrid STFT and 2D2PCA (STFT2D 2 PCA) method.…”