Neural interfacing with the peripheral nervous system

Abstract

Although electrical stimulation has proven to be capable of restoring neuronal function in the damaged or injured nervous system, there are several limitations to this technique. The availability of electrodes capable of selective fascicle recruitment and physiological fiber diameter recruitment (from small to large) is crucial for the development of successful prostheses. Nerve cuff electrodes have several advantages over other methods since they can provide activation of multiple muscles groups from a single site. Current cuff electrodes are reshaping the nerve into round shapes and making it difficult to recruit selectively fibers in the center of the nerve. Yet two major problems have not yet found satisfactory solutions: 1) fascicle selectivity and fiber diameter selectivity. We present here a new design that reshapes the nerve into a flat configuration, the flat interface nerve electrode (FINE). This design can improve the ability of the electrode to selectively activate the various fascicles of the nerve. Experiments to measure this selectivity were carried out on the hypoglossal nerve and its three main branches. The ability to recruit various fascicles was estimated using a selectivity index (SI). The overall performance of the FINE, as defined by the selectivity index (SI), showed a high degree of selectivity at both the fascicular and muscular levels: 0.91 +/- 0.05 (n = 5) and 0.85 +/- 0.03 (n = 4), respectively. This flat interface design minimizes the maximum distance between each contact and the fibers. Computer simulation have shown that it is possible to reverse the recruitment order by using electrode arrays placed along the nerve. This model prediction was tested in the lateral gastrocnemius/soleus branch of the sciatic nerve in cats since these muscles are innervated by fibers with different diameters. A stimulus electrode was placed around LG nerve. Tendons of LG and soleus muscles were separated and attached to two independent force transducers. The recruitment curves generated by tripolar and array electrodes were compared. Tripolar stimulation recruited LG before soleus muscles as expected, whereas the electrode array fully activated soleus while activating only 50% of LG muscles. These results show that the electrode - array is capable of reversing the recruitment order by manipulating the extracellular voltage along the nerve.