Toward the Next Generation of Retinal Neuroprosthesis: Visual Computation with Spikes

Abstract

Neuroprosthesis, as one type of precision medicine device, is aiming for manipulating neuronal signals of the brain in a closed-loop fashion, together with receiving stimulus from the environment and controlling some part of our brain/body. In terms of vision, incoming information can be processed by the brain in millisecond interval. The retina computes visual scenes and then sends its output as neuronal spikes to the cortex for further computation. Therefore, the neuronal signal of interest for retinal neuroprosthesis is spike. Closed-loop computation in neuroprosthesis includes two stages: encoding stimulus to neuronal signal, and decoding it into stimulus. Here we review some of the recent progress about visual computation models that use spikes for analyzing natural scenes, including static images and dynamic movies. We hypothesize that for a better understanding of computational principles in the retina, one needs a hypercircuit view of the retina, in which different functional network motifs revealed in the cortex neuronal network should be taken into consideration for the retina. Different building blocks of the retina, including a diversity of cell types and synaptic connections, either chemical synapses or electrical synapses (gap junctions), make the retina an ideal neuronal network to adapt the computational techniques developed in artificial intelligence for modeling of encoding/decoding visual scenes. Altogether, one needs a systems approach of visual computation with spikes to advance the next generation of retinal neuroprosthesis as an artificial visual system.

Reconstruction of natural visual scenes from neural spikes with deep neural networks

“…The retina has been thought of as a relatively simple neuronal circuit, however, it has been suggested that many complex computations can be realized by their neurons, in particular, the RGCs [11]. It has been studied intensively to develop various types of models for the encoding of visual scenes in the retina [1,13,16]. Similar to our current study, other recent work also focuses on neuronal spikes in the retina to develop the novel decoding methods [17,24,25].…”

“…Various neuroscience mech-anisms have been identified in understanding the retinal computation of visual scenes by its neurons and neural circuitry [6,7,8,9,10,11]. In addition, for understanding the encoding principles of the retina, a number of models are developed based on different properties of neurons and neural circuits in the retina [12,13,14,15,16].…”

Probabilistic inference of binary Markov random fields in spiking neural networks through mean-field approximation

“…As discussed in [32], we can assume that the actual firing ratê r i (t) is linearly related to the "firing rate" r i (t) obtained from Eq. (16), that is,…”

“…Or more precisely, how a network of spiking neurons in the brain can implement inference of probabilistic graphical models? This problem is of great importance to both computer science and brain science [16]. If we known the neural algorithms of probabilistic inference, it is possible to build a machine that can perform probabilistic inference like the human brain.…”

Neural System Identification With Spike-Triggered Non-Negative Matrix Factorization

“…Even for the retina, a relatively simple neuronal circuit, the underlying structure and, in particular, its functional characteristics are still not completely understood. However, the retina serves as a typical model for both deciphering the structure of neuronal circuits [1], [2], [3], [4], [5], [6] and testing novel methods for neuronal coding [7], [8], [9], [10], [11]. The retina consists of three layers with photoreceptors, bipolar cells, and ganglion cells together with inhibitory horizontal and amacrine cells in between as illustrated in Fig.…”