Evaluation of Hand Action Classification Performance Using Machine Learning Based on Signals from Two sEMG Electrodes.

Classification-based myoelectric control has attracted significant interest in recent years, leading to prosthetic hands with advanced functionality, such as multi-grip hands. Thus far, high classification accuracies have been achieved by increasing the number of surface electromyography (sEMG) electrodes or adding other sensing mechanisms. While many prescribed myoelectric hands still adopt two-electrode sEMG systems, detailed studies on signal processing and classification performance are still lacking. In this study, nine able-bodied participants were recruited to perform six typical hand actions, from which sEMG signals from two electrodes were acquired using a Delsys Trigno Research+ acquisition system. Signal processing and machine learning algorithms, specifically, linear discriminant analysis (LDA), k-nearest neighbors (KNN), and support vector machines (SVM), were used to study classification accuracies. Overall classification accuracy of 93 ± 2%, action-specific accuracy of 97 ± 2%, and F1-score of 87 ± 7% were achieved, which are comparable with those reported from multi-electrode systems. The highest accuracies were achieved using SVM algorithm compared to LDA and KNN algorithms. A logarithmic relationship between classification accuracy and number of features was revealed, which plateaued at five features. These comprehensive findings may potentially contribute to signal processing and machine learning strategies for commonly prescribed myoelectric hand systems with two sEMG electrodes to further improve functionality.

Assessment of 3D printed mechanical metamaterials for prosthetic liners.

This study focuses on novel design and evaluation of Elastic 50A (EL50) mechanical metamaterials with open-cell patterns for its potential application to lower limb residuum/socket interfaces, specifically that of a transtibial (TT) amputee. Mechanical characteristics, that is, effective Young's modulus (E), was tuned by altering metamaterial porosity, which was experimentally verified. Specifically, pore radius of the unit cell was varied to achieve a range of E-values (0.05-1.71 MPa) for these 3D printed metamaterials. Finite Element Analysis (FEA) was conducted to evaluate pressure distribution across key load-bearing anatomical sites of a TT residuum. Using designed metamaterials for homogeneous liners, pressure profiles were studied and compared with a silicone liner case. Additionally, a custom metamaterial liner was designed by assigning appropriate metamaterials to four load-sensitive and tolerant anatomical sites of the TT residuum. The results suggest that lowest pressure variation (PV), as a measure of pressure distribution levels and potential comfort for amputees, was achieved by the custom metamaterial liner compared to any of the homogeneous liners included in this study. It is envisaged that this work may aid future design and development of custom liners using now commonly available 3D printing technologies and available elastomer materials to maximise comfort, tissue safety and overall rehabilitation outcomes for lower limb amputees.

Assessing Socket Fit Effects on Pressure and Shear at a Transtibial Residuum/Socket Interface.

Fluctuations in residuum volume during daily activities are known to occur in lower-limb amputees. This can cause frequent changes to fit, which cannot be accommodated by commonly-used prosthetic sockets. The real-time effects, if any, of these minor socket fit changes on interface biomechanics have not been studied extensively. Amputees commonly use different layers of socks to accommodate frequent volume fluctuations, enabling adjustment of socket fit. We, thus, altered socket fit levels via addition/removal of sock layers to a transtibial amputee who habitually-donned two-sock layers to mimic relatively looser and tighter socket fits. Interface pressure and shear sensors were placed at known prominent load-bearing sites of the transtibial residuum/socket interface, i.e., patellar tendon (PT), popliteal fossa (PF), and anterior-distal (AD) end, to measure real-time biomechanical interactions during standing and level walking. Although socket fit level was only slightly modified, changes in interface pressure and shear across anatomical sites were still observed. Tighter fit corresponds to notable pressure reduction at AD during early stance and pressure increase at PT during terminal stance due to the residuum being pushed up. Shear-to-pressure ratios were used to assess comfort, while pressure- and shear-time integrals were used to assess tissue health. We observed more notable changes at tissue sites (e.g., AD and PF). Combined evaluation of pressure and shear, including shear-to-pressure ratio and time integrals, may offer insight for residuum care.