Artificial neural net based signal processing for interaction with peripheral nervous system

Two artificial neural Net (ANN) based signal processing systems processing signals using interfaces to the peripheral nervous system are presented. The aim of the paper is to show ANN´s capability to meet requirements needed to interact with the biological nervous system. First, a system for classification of nerve signals is presented. Recordings of nerve signals done by regeneration type neurosensors interfacing the peripheral nerve system are processed by ANNs in order to identify their origin axons from a recorded mixture of several axons. The second system presented is somehow the inverse of the processing system above. In this case, the aim of the signal processing system is to introduce information to the peripheral nervous system by computing appropriate stimulus pattern for functional electrical stimulation. The connection to the peripheral nervous system is done by Cuff-electrodes.