AbstractsComputer Science

Closed-loop optimization of extracellular electrical stimulation for targeted neuronal activation

by Michelle Lea Kuykendal




Institution: Georgia Tech
Department: Electrical and Computer Engineering
Degree: PhD
Year: 2014
Keywords: Selective stimulation; Extracellular electrical stimulation; MEA; Microelectrode array; Optical recording; Stochastic response; Probabilistic activation
Record ID: 2042975
Full text PDF: http://hdl.handle.net/1853/52303


Abstract

We have developed a high-throughput system of closed-loop electrical stimulation and optical recording that facilitates the rapid characterization of extracellular stimulus-evoked neural activity. The ability to selectively stimulate a neuron is a defining characteristic of next-generation neural prostheses. Greater stimulus control and differential activation of specific neuronal populations allows for prostheses that better mimic their biological counterparts. In our system, we deliver square current pulses using a microelectrode array; automated real-time image processing of high-speed digital video identifies the neuronal response; and a feedback controller alters the applied stimulus to achieve a targeted response. The system controller performs directed searches within the strength-duration (SD) stimulus parameter space to build probabilistic neuronal activation curves. An important feature of this closed-loop system is a reduction in the number of stimuli needed to derive the activation curves when compared to the more commonly used open-loop system: this allows the closed-loop system to spend more time probing stimulus regions of interest in the multi-parameter waveform space, facilitating high resolution analysis. The stimulus-evoked activation data were well-fit to a sigmoid model in both the stimulus strength (current) and duration (pulse width) slices through the waveform space. The 2-D analysis produced a set of probability isoclines corresponding to each neuron-electrode pairing, which were fit to the SD threshold model described by Lapique (1907). We show that stimulus selectivity within a given neuron pair is possible in the one-parameter search space by using multiple stimulation electrodes. Additionally, by applying simultaneous stimuli to adjacent electrodes, the interaction between stimuli alters the neuronal activation threshold. The interaction between simultaneous multi-electrode multi-parameter stimulus waveforms creates an opportunity for increased stimulus selectivity within a population. We demonstrated that closed-loop imaging and micro-stimulation technology enable the study of neuronal excitation across a large parameter space, which is requisite for controlling neuronal activation in next generation clinical solutions.