Comparing Implantable Epimysial and Intramuscular Electrodes for Prosthetic Control

Eric J Earley^1,2,3,4*Morten Bak Kristoffersen^3,5Max Ortiz-Catalan^2,3,6,7

• ¹Department of Orthopedics, University of Colorado Anschutz Medical Campus, Aurora, United States
• ²Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden
• ³Center for Bionics and Pain Research, Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden
• ⁴Bone-Anchored Limb Research Group, University of Colorado Anschutz Medical Campus, Aurora, United States
• ⁵Department of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Mölndal, Sweden
• ⁶Prometei Pain Rehabilitation Center, Vinnytsia, Ukraine
• ⁷Center for Complex Endoprosthetics, Osseointegration, and Bionics, Kyiv, Ukraine

Implantable electrodes are the subject of increasing interest due to the possibilities they present for the control of assistive devices such as prosthetic limbs, however evidence as to whether epimysial or intramuscular electrodes result in better performance is lacking. In this retrospective study, we analyzed data collected from six users of a neuromusculoskeletal arm prosthesis who had been implanted with epimysial or intramuscular electrodes, two of whom were implanted with both electrode types, and compared electrical impedance and electromyographic feature space characteristics -all important qualities for the control of prosthetic arms and hands. Our results showed significantly greater cross-channel impedances for intramuscular electrodes suggesting improved isolation and reduced cross-talk, however this did not translate to significantly improved signal-to-noise ratio, separability, or repeatability. Sequential feedforward selection analysis may suggest that epimysial electrodes contribute greater signal separability when recording from larger muscles used for gross hand movements, whereas intramuscular electrodes contribute greater signal separability when recording from smaller muscles used for grasp prehension and finger movements, but additional study is required to confirm these findings. Our results provide preliminary understanding as to which electrodes should be used for which patients, which may help to guide clinical practice for future implementation of cutting-edge bionic arms.