CAREER: Low-Power VLSI Circuits for Large-Scale Neuronal Recording

University Of Utah, Salt Lake City UT

Investigators

Abstract

There is a great demand for technologies that enable neuroscientists and clinicians to observe the simultaneous activity of large numbers of neurons in the brain. The monitoring of these groups or "neural ensembles" allows researchers to begin to understand the cooperative mechanisms used by neurons to encode and process information. Recent advances in MEMS technology have produced small arrays of microelectrodes containing as many as 100 recording sites. "Next generation" neural recording systems must be capable of observing 100-1000 neurons simultaneously, in a fully-implanted unit. While integrated electronics have been developed for small-scale amplification of the weak extracellular neural signals (<100 electrodes), existing circuits have high levels of noise and consume too much power to be fully implanted in larger quantities. We propose to develop low-power, low-noise analog and mixed-signal VLSI systems allowing fully implantable recording of 100-1000 neurons. A fully implanted multichannel neural recording system must use an RF or inductive-link transmitter for transcutaneous telemetry. We will investigate techniques for on-chip data reduction (e.g., spike thresholding, feature detection) to assist in spike sorting and reduce the required bandwidth (and hence power) of such a transmitter. The educational component of the proposed work involves the improvement of the VLSI curriculum in the PI's department. This improvement will consist of three main thrusts: (1) Development of a laboratory component of a course in analog integrated circuit design taught by the PI. The construction of "class chips" will allow students to measure VLSI circuits in modern submicron fabrication technologies. (2) Development of a new advanced analog VLSI course. (3) Enlisting industrial partners to evaluate our VLSI curriculum. In addition to this curriculum development, the PI will also mentor a graduate student who will perform research related to the proposal.