Back to Annual Meeting Posters
Myoelectric Signal Transmission from Implanted Epimysial Electrodes Using a Bone-Anchor as a Conduit
Yazan Al-Ajam, MBChB, MRCS1; Henry Lancashire, BSc, MRes1; Catherine Pendegrass, PhD1; Norbert Kang, MBBS, MD, FRCS2; Robert P. Dowling, BSc, (Hons)1; Steven J.G. Taylor, PhD1; Gordon Blunn, PhD1
1University College London, London, United Kingdom; 2Royal Free Hospital, London, United Kingdom
Introduction: Current Upper-limb myoelectric prostheses rely on only 2 control signals from surface electrodes, placed over antagonistic muscles in the amputation stump, for limb control. While this has benefits over the traditional body-powered control, there are disadvantages; electrode lift-off, impedance variation, cross-talk, reliability, and limitations in intuitive control. To address these problems, electrodes can be implanted directly on individual muscles responsible for specific actions. Not only does this address skin-related issues, it reduces cross-talk and greatly increases the number of control channels for multi degrees of freedom intuitive control. Bone-anchored devices can be used to overcome problems with prosthetic attachment and additionally used to transfer control signals from these implantable electrodes to the prosthesis.  In above-elbow amputees, targeted muscle reinnervation (TMR) enables more signal generation by redirecting nerves previously controlling the amputated muscles in the forearm, to surrogate muscles in the torso (e.g. pectoralis major).  We describe in vivo model using implantable electrodes to record myoelectric signals (MES) in normal muscles and following TMR, utilizing a bone-anchor as a conduit to carry signals across the skin barrier.
Materials and Methods: An in vivo n=6 ovine model was used. A bone-anchor was placed trans-tibially and bipolar electrodes sutured to M. Peroneus Tertius (PT). In a further n=1, motor nerve to PT was divided and coapted with a motor branch from peroneal nerve. MES were recorded over a 12-week period. Functional recovery in the TMR model was assessed by MES and force-plate analysis (FPA).
Results: In the n=6 group, there was a positive correlation between signal to noise ratio (SNR) and time since implantation (p < 0.005), with a mean SNR of 7 by week 12. In the TMR model, functional recovery was observed after 6 weeks. Difference between legs returned to normal (pre-op: left 4.7 N/kg, right 4.8 N/kg; 80 days post-op: left 4.1 N/kg, right 4.3 N/kg). Recorded MES from TMR muscle compared favourably with healthy muscle.
Conclusions: We have demonstrated that a bone-anchor is a reliable and robust conduit for transmitting MES over a period of 12 weeks. The combination of implanted electrodes & direct skeletal fixation offers clear advantages over current systems for prosthetic attachment & control. This system forms the basis of a complete solution for prosthetic rehabilitation, which can also be used in the context of TMR.
References: 1. Al Ajam et. al., 2013. PMID: 23358938
2. Kuiken et. al., 2004. PMID: 15658637
Back to Annual Meeting Posters