A database of high-density surface electromyogram signals comprising 65 isometric hand gestures.

Control of contemporary, multi-joint prosthetic hands is commonly realized by using electromyographic signals from the muscles remaining after amputation at the forearm level. Although this principle is trying to imitate the natural control structure where muscles control the joints of the hand, in practice, myoelectric control provides only basic hand functions to an amputee using a dexterous prosthesis. This study aims to provide an annotated database of high-density surface electromyographic signals to aid the efforts of designing robust and versatile electromyographic control interfaces for prosthetic hands. The electromyographic signals were recorded using 128 channels within two electrode grids positioned on the forearms of 20 able-bodied volunteers. The participants performed 65 different hand gestures in an isometric manner. The hand movements were strictly timed using an automated recording protocol which also synchronously recorded the electromyographic signals and hand joint forces. To assess the quality of the recorded signals several quantitative assessments were performed, such as frequency content analysis, channel crosstalk, and the detection of poor skin-electrode contacts.

Extraction of Multi-Labelled Movement Information from the Raw HD-sEMG Image with Time-Domain Depth.

In contemporary muscle-computer interfaces for upper limb prosthetics there is often a trade-off between control robustness and range of executable movements. As a very low movement error rate is necessary in practical applications, this often results in a quite severe limitation of controllability; a problem growing ever more salient as the mechanical sophistication of multifunctional myoelectric prostheses continues to improve. A possible remedy for this could come from the use of multi-label machine learning methods, where complex movements can be expressed as the superposition of several simpler movements. Here, we investigate this claim by applying a multi-labeled classification scheme in the form of a deep convolutional neural network (CNN) to high density surface electromyography (HD-sEMG) recordings. We use 16 independent labels to model the movements of the hand and forearm state, representing its major degrees of freedom. By training the neural network on 16 × 8 sEMG image sequences 24 samples long with a sampling rate of 2048 Hz to detect these labels, we achieved a mean exact match rate of 78.7% and a mean Hamming loss of 2.9% across 14 healthy test subjects. With this, we demonstrate the feasibility of highly versatile and responsive sEMG control interfaces without loss of accuracy.