Electronic bidirectional interfaces to the peripheral nervous system for prosthetic applications

The research presented in this thesis concerns the field of bioelectronics, in particular the work has been focused on the development of special electronic devices for neural signal acquisition and Peripheral Nervous System (PNS) stimulation. The final aim of the project in which this work is involved is in fact the realization of a prosthetic hand controlled using neural signals. The commercially available prosthesis are based on Electromyographic (EMG) signals, their use implies unnatural movements for the patient that needs a special training to develop the control capabilities over the mechanical limb. The proposed approach offers a number of advantages compared to the traditional prosthesis, first because the signals used are the same used to control the biologic limb, allowing a more comfortable solution for the patient that gets closer to feel the robotic hand as a natural extension of his/her body. Secondly, placing temperature and pressure sensors on the limb surface, it is possible to trasduce such information in an electrical current that, injected into the PNS, can restore the sensory feedback in amputees. The final goal of this research is the development of a fully implantable device able to perform a bidirectional communication between the robotic hand and the patient. Due to small area, low noise and low power constraints, the only possible way to reach this aim is the design of a full custom Integrated Circuit (IC). However a preliminary evaluation of the key design features, such as neural signal amplitudes and frequencies as well as stimulation shape parameters, is necessary in order to define clearly and precisely the design specifications. A low-cost and short implementation time device is then needed for this aim, the Components Off The Shelf (COTS) approach seems to be the best solution for this purpose. A Printed Circuit Board (PCB) with discrete components has been designed, developed and tested, the information extracted by the test results have been used to guide the IC design. The generation of electrical signals in biological cells, such as neural spikes, is possible thanks to ions that move across the cell membrane. In many applications it is important, not only to record the spikes, but also to measure these small currents in order to understand which electro-chemical processes are involved in the signal generation and to have a direct measurement of the ion channels involved in the reaction. Ion currents, in fact, play a key role in several physiological processes, in neural signal generation, but also in the maintenance of heartbeat and in muscle contraction. For this purpose, a system level implementation of a Read out circuit for ion channel current detection has been developed.