Optimizing nerve cuff stimulation of targeted regions through use of genetic algorithms

Abstract

A nerve cuff electrode is a viable technology for use in a neuroprostheses system to restore loss of function due to neurological injury. The Flat Interface Nerve Electrode (FINE) is a nerve cuff that gently reshapes the nerve to bring the axons closer to the stimulating contacts. The overall goal of this work is to optimize nerve cuff stimulation in upper extremity nerves. Recently, highly efficient and accurate linear models of neuronal activation have been developed in our lab. Using the fast calculations from the newly developed linear activation method, nerve stimulation parameters such as current pulse width and pulse amplitude at many electrode contacts can be explored by employing optimization algorithms. Finite element nerve models with high density electrodes were constructed based on upper extremity cadaveric nerve cross sections. An objective function was developed to target specific groups of nerve fascicles and minimize overlap amongst these groups. By changing the objective function and using a genetic search algorithm, stimulation parameters can be optimized for many contacts.