Computational modelling of the suppression of optic nerve fibre

Retinal neuroprostheses aim to restore vision in degenerated retina but are hindered by poor activation selectivity. Suppressing subsets of optic nerve fibres based on diameter may improve selectivity by filtering superfluous activations. This study evaluates frequency-induced neuromodulation (FIN) using sinusoidal waveforms to achieve diameter-selective suppression. Computational models of retinal ganglion cells (RGCs) and optic nerve fibres (1.4, 2.8, and 4.3 μm) were subjected to sinusoidal currents from 10 to 10,000 Hz. The range of optic nerve fibre diameters was chosen to represent the optic nerve fibre population in rats. FIN could selectively filter action potential conduction, with the highest suppression probability at frequencies of 1000 Hz and above. The current amplitude required for suppression was influenced by the diameter of the fibres, the sinusoidal frequency of the FIN waveforms, and the ON-OFF typing of the RGC. It was observed that the membrane potential upstream of the FIN application site influenced the response to FIN. Selective suppression based on the fibre diameter is plausible, whereas selective functional-suppression based on the ON-OFF typing is less plausible owing to the small differences in their respective suppression thresholds. Filtering of this kind may improve the utility of retinal neuromodulation as a treatment to blindness.