Inkjet-printed fully customizable and low-cost electrodes matrix for gesture recognition

Abstract

The use of surface electromyography (sEMG) is rapidly spreading, from robotic prostheses and muscle computer interfaces to rehabilitation devices controlled by residual muscular activities. In this context, sEMG-based gesture recognition plays an enabling role in controlling prosthetics and devices in real-life settings. Our work aimed at developing a low-cost, print-and-play platform to acquire and analyse sEMG signals that can be arranged in a fully customized way, depending on the application and the users’ needs. We produced 8-channel sEMG matrices to measure the muscular activity of the forearm using innovative nanoparticle-based inks to print the sensors embedded into each matrix using a commercial inkjet printer. Then, we acquired the multi-channel sEMG data from 12 participants while repeatedly performing twelve standard finger movements (six extensions and six flexions). Our results showed that inkjet printing-based sEMG signals ensured significant similarity values across repetitions in every participant, a large enough difference between movements (dissimilarity index above 0.2), and an overall classification accuracy of 93–95% for flexion and extension, respectively.

Multi‐Electrode Printed Bioelectronic Patches for Long‐Term Electrophysiological Monitoring

“…Figure S13A (Supporting Information) shows the dissimilarity coefficient between gesture pairs (computed using the Spearman correlation‐based dissimilarity coefficient, i.e., the spearman distance between feature vectors of each pose observation, averaged across three one‐second windows). While high dissimilarity values between pairs of gestures indicate an easy distinction between those poses (as is the case of the pair six and 12), dissimilarity coefficients as low as 0.1 [ 70 ] or 0.2 [ 71 ] have previously shown good results in automatic EMG hand gesture classifiers. Overall, 85% of the gesture pairs show a dissimilarity coefficient >0.2, while the average dissimilarity is 0.342.…”

“…A similar work using an equivalent EMG electrode montage [ 71 ] has been able to distinguish only between six hand gestures with a classification accuracy of ≈95% on 12 participants. The present work improves these results not only in terms of correlation coefficients but also due to the use of a considerably simpler (in terms of features extracted from the EMG signal) and more wearable system.…”

A Decade of Progress in Human Motion Recognition: A Comprehensive Survey From 2010 to 2020

“…The system captured EMG signals from the forearm muscles and extracted features using a time-domain analysis. In contrast, Rosati et al [116] presented a low-cost, inkjetprinted electrode matrix for gesture recognition. The system uses capacitive sensing with a custom-made printed circuit board and achieves an average recognition accuracy of 88.3% for seven hand gestures.…”

3D Printed Electromyography Sensing Systems

“…The PLA electrodes performed the gesture classification with an average of 85% accuracy. In another study, Rosati et al [73] approached gesture recognition with the printed electrode matrix structure. The electrode consists of eight channels and is positioned on the forearm for EMG detection.…”