A VLSI Field-Programmable Mixed-Signal Array to Perform Neural Signal Processing and Neural Modeling in a Prosthetic System

Abstract

A very-large-scale integration field-programmable mixed-signal array specialized for neural signal processing and neural modeling has been designed. This has been fabricated as a core on a chip prototype intended for use in an implantable closed-loop prosthetic system aimed at rehabilitation of the learning of a discrete motor response. The chosen experimental context is cerebellar classical conditioning of the eye-blink response. The programmable system is based on the intimate mixing of switched capacitor analog techniques with low speed digital computation; power saving innovations within this framework are presented. The utility of the system is demonstrated by the implementation of a motor classical conditioning model applied to eye-blink conditioning in real time with associated neural signal processing. Paired conditioned and unconditioned stimuli were repeatedly presented to an anesthetized rat and recordings were taken simultaneously from two precerebellar nuclei. These paired stimuli were detected in real time from this multichannel data. This resulted in the acquisition of a trigger for a well-timed conditioned eye-blink response, and repetition of unpaired trials constructed from the same data led to the extinction of the conditioned response trigger, compatible with natural cerebellar learning in awake animals.

A Cerebellar Neuroprosthetic System: Computational Architecture and in vivo Test

“…Here, we emphasize steps 1–3 since we already previously have demonstrated step 4 (Hofstotter et al, 2004; Bamford et al, 2012). …”

“…However, during a CS presentation, Purkinje cells switch to a decaying activity. For a detailed explanation see Hofstotter et al (2004). …”

“…Some examples are retinal and cochlear implants (Eddington et al, 1978; Wilson et al, 1991; Zrenner, 2002; Cohen, 2007), and brain computer interface systems controlling artificial limbs (Chapin et al, 1999; Schwartz et al, 2006; Moritz et al, 2008; Hochberg et al, 2012). However, the bi-directional coupling of a prosthetic system with the central nervous system has only very recently been addressed (Berger et al, 2011; Bamford et al, 2012; Giovannucci et al, 2010). Here we demonstrate the functional bi-directional coupling of an artificial system and the central nervous system in the context of classical conditioning.…”

“…Based on these assumptions, we have previously developed a computational model of learning in the cerebellum (Verschure and Mintz, 2001; Hofstotter et al, 2002) that was implemented in a VLSI hardware and tested in a robot learning task (Hofstotter et al, 2004). Here, we show how an updated version of this computational model can be deployed as a prototype of a neuroprosthetic device and interfaced with the brain of a living animal, replacing the function of the animal’s inactivated cerebellum.…”

“…Indeed, the features of the experimental recordings presented here match the known IO physiology, showing a baseline firing rate of 0.5–2 Hz (De Zeeuw et al, 1998). Therefore, the encoding of the US in the IO will be very noisy, and will prevent learning if we naïvely use the model in Hofstotter et al (2004). Hence, our first goal is to update such model in order to cope with biosignals.…”