Neural Prostheses

Sensory Recovery & Plasticity

Millions of people worldwide have disabilities due to spinal cord and peripheral nerve injuries. One approach for the treatment of these patients is the use of neuroprostheses, where brain or nerve signals interact with an implanted device to control motion and convey sensory information. Great advances have been made to allow patients to control robotic limbs using decoded brain signals, peripheral nerve signals, or muscle signals. However, the sensory feedback component of these devices has been largely ignored. The lack of tactile sensory feedback on the hand inevitably affects the prehensile function of the hand, such as lifting or holding objects in daily activities. Fine touch feedback from the fingertips is necessary for precision tasks like the manipulation of small objects, such as using coins for paying.

Overview of final implemented system.
The complete system in human patients: the Silicodermis glove receives sensory input, which is converted into neuromorphic stimulation by the Neurointerpreter. Stimulation commands are transmitted wirelessly to the implanted Stimulation unit to stimulate the peripheral nerves using the Neuroport neural interface. Our laboratory is designing an integrated system consisting of a neuroprosthetic glove with temperature and pressure sensors, and associated circuitry that will translate the sensory data to biomimetic patterns of electrical stimulation. We also are designing a neuromorphic stimulator integrated circuit (IC) with an integrated neural interface that will wirelessly stimulate the peripheral nerve


Glove design illustrating three basic movements using joint angle sensors.
Fine object discrimination and touch will be decoded based on the grid of tactile sensors on one of the glove's fingertip.
For our sensory approach, we will rely on the plasticity of the brain to interpret new sensory input, via stimluation, that corresponds to the sensation at the tactile sensory within the cybernetic glove. To this end, the patient will need to calibrate the device by providing user feedback. I.e., what sensation do you feel when such a stimulation is applied? The correct tactile sensors can then be matched with the user feedback, as well as stimulation parameters programmed in the microcontroller of the implanted device.

Figures and text from these pages may not be reproduced or used for any purpose without permission from SINAPSE.