A Gaussian Mixture Model Based Classification Scheme for Myoelectric Control of Powered Upper Limb Prostheses

Abstract

This paper introduces and evaluates the use of Gaussian mixture models (GMMs) for multiple limb motion classification using continuous myoelectric signals. The focus of this work is to optimize the configuration of this classification scheme. To that end, a complete experimental evaluation of this system is conducted on a 12 subject database. The experiments examine the GMMs algorithmic issues including the model order selection and variance limiting, the segmentation of the data, and various feature sets including time-domain features and autoregressive features. The benefits of postprocessing the results using a majority vote rule are demonstrated. The performance of the GMM is compared to three commonly used classifiers: a linear discriminant analysis, a linear perceptron network, and a multilayer perceptron neural network. The GMM-based limb motion classification system demonstrates exceptional classification accuracy and results in a robust method of motion classification with low computational load.

In-Home Assistive System for Traumatic Brain Injury Patients

“…As indicated previously these errors occur mostly in the transition between tasks: the hand gestures related to these segments are very similar, such as grabbing the shaver or tooth brush. Similar problems were observed in the classification of EMG signals in [5]. The authors have noticed that the GMM based system performed poorly in the transitions from one activity to another.…”

“…By tuning the MV with thresholds, we observed a significant decrease in the number of FPs, which occurred in the recognition of almost every trial. The same approach may also be a solution for the transition problems defined in [5]. The computational complexity of the algorithms is low and can be installed into the wireless sensor kits in order to develop an intelligent sensor network.…”

“…3(b). GMM is widely used in continuous classification of EMG signals for prosthetic control and speaker identification problems due to their robustness and lower computational complexity [5,6]. A GMM probability density function (pdf), is defined as a weighted combination of N Gaussians.…”

sEMG based quantitative assessment of acupuncture on Bell’s palsy: an experimental study

“…According to (5), we set T = 0.5 ms, T w = 128 ms, T d = 32 ms in our experiments. The selection of these parameters is weighted in [27,28]. The sampled sEMG data with respect to each patient are respectively segmented to form overlapping data sequences for healthy and diseased sides.…”

Virtual Control of Prosthetic Hand Based on Grasping Patterns and Estimated Force from Semg

“…This paper, based on Virtual Reality Modeling Language(VRML) [6,7] and Simulink 3D Animation, establishes a virtual environment for multifreedom prosthetic hand.…”