Dimensionality reduction for classification of object weight from electromyography

Electromyography (EMG) is a simple, non-invasive, and cost-effective technology for measuring muscle activity. However, multi-muscle EMG is also a noisy, complex, and high-dimensional signal. It has nevertheless been widely used in a host of human-machine-interface applications (electrical wheelchairs, virtual computer mice, prosthesis, robotic fingers, etc.) and, in particular, to measure the reach-and-grasp motions of the human hand. Here, we developed an automated pipeline to predict object weight in a reach-grasp-lift task from an open dataset, relying only on EMG data. In doing so, we shifted the focus from manual feature-engineering to automated feature-extraction by using pre-processed EMG signals and thus letting the algorithms select the features. We further compared intrinsic EMG features, derived from several dimensionality-reduction methods, and then ran several classification algorithms on these low-dimensional representations. We found that the Laplacian Eigenmap algorithm generally outperformed other dimensionality-reduction methods. What is more, optimal classification accuracy was achieved using a combination of Laplacian Eigenmaps (simple-minded) and k-Nearest Neighbors (88% F1 score for 3-way classification). Our results, using EMG alone, are comparable to other researchers’, who used EMG and EEG together, in the literature. A running-window analysis further suggests that our method captures information in the EMG signal quickly and remains stable throughout the time that subjects grasp and move the object.

On the Classification of ECG and EEG Signals with Various Degrees of Dimensionality Reduction

Classification performances for some classes of electrocardiographic (ECG) and electroencephalographic (EEG) signals processed to dimensionality reduction with different degrees are investigated. Results got with various classification methods are given and discussed. So far we investigated three techniques for reducing dimensionality: Laplacian eigenmaps (LE), locality preserving projections (LPP) and compressed sensing (CS). The first two methods are related to manifold learning while the third addresses signal acquisition and reconstruction from random projections under the supposition of signal sparsity. Our aim is to evaluate the benefits and drawbacks of various methods and to find to what extent they can be considered remarkable. The assessment of the effect of dimensionality decrease was made by considering the classification rates for the processed biosignals in the new spaces. Besides, the classification accuracies of the initial input data were evaluated with respect to the corresponding accuracies in the new spaces using different classifiers.

Improved weighted local linear embedding algorithm based on Laplacian eigenmaps

To solve the shortcomings of local linear embedding (LLE), such as sensitive to noise and poor generalization ability for new samples, an improved weighted local linear embedding algorithm based on Laplacian eigenmaps (IWLLE-LE) is proposed in this paper. In the proposed algorithm, Laplacian eigenmaps are used to reconstruct the objective function of dimensionality reduction. The weights of it are introduced by combining the geodesic distance with Euclidean distance, which can effectively represent the manifold structure of nonlinear data. Compared the existing LLE algorithm, the proposed one better maintains the original manifold structure of the data. The merit of the proposal is enhanced by the theoretical analysis and numerical experiments, where the classification recognition rate is 2%–8% higher than LLE.

The Laplacian eigenmaps dimensionality reduction of fMRI data for discovering stimulus-induced changes in the resting-state brain activity

The resting brain at wakefulness is active even in the absence of goal-directed behavior or salient stimuli. However, patterns of this resting-state (RS) activity can undergo alterations following exposure to meaningful stimuli. This study aimed to develop an unbiased method to detect such changes in the RS activity after exposure to emotionally meaningful stimuli. For this purpose, we used functional magnetic resonance imaging (fMRI) of RS brain activity before and after the acquisition and extinction of experimental conditioned fear. A group of healthy volunteers participated in three fMRI sessions: a RS before fear conditioning, a fear extinction session, and a RS immediately after fear extinction. The fear-conditioning paradigm consisted of three neutral visual stimuli paired with a partial reinforcement by a mild electric current. We used both linear and non-linear dimensionality reduction approaches to distinguish between the initial RS and the RS after stimuli exposure. The principal component analysis (PCA) as a linear dimensionality reduction method did not differentiate these states. Using the non-linear Laplacian eigenmaps manifold learning method, we were able to show significant differences between the two RSs at the level of individual participants. This detection was further improved by smoothing the BOLD signal with the wavelet multiresolution analysis. The developed method can improve the discrimination of functional states collected in longitudinal fMRI studies.