Investigation of a Novel Strategy to Promote Enhanced Visual Sensations by Electrical Retinal Implants by Means of Alternating Topological Activation of the Stimulation Electrodes

Abstract

Multielectrode Electrical (E-) stimulation (stim) of the retina is used to restore vision in blind patients suffering from photoreceptor dystrophy. In recent clinical trials, this approach has shown to be beneficial for blind patients to perform daily life tasks. However, the visual perceptions of the patients are limited in spatial and temporal resolution. Repeated E-stim of the retina leads to image fading at E-stim frequencies higher than ~5 Hz. The main focus of this thesis was to investigate a novel E-stim strategy to maintain local visual responses, thereby reducing image fading effects at high E-stim frequency. An electrophysiological set-up with a hexagonal multielectrode array (MEA) was used for experiments on explanted healthy mouse retinas. Preliminary light (L-) stim experiments were conducted to observe the responses induced by a bright Square moving in human microsaccade-like (hMS -like) moves as opposed to the responses induced due to stationary presentation of the Square. The stationary presentation of a Square resulted in a slow reduction of responses, presumably leading to image-fading; if the bright Square moved along the hMS-like sequence, each movement resulted in reoccurring of responses presumably leading to image-refreshing. Initial E- stim experiments were conducted to obtain tuning curves based on the strength of the stimulus. A monomodal cathodic stimulation strength of -1000 mV with a pulse duration of 0.5 ms was used for further experiments. Subsequently, a hexagon electrical pattern with 80 μm diagonal extension was created using 6 MEA-electrodes (6-electrode-hexagon, 6-e-hexagon). The 6-e-hexagon represented an approximate pitch on the Alpha IMS and AMS E-retinal implants (70 μm) used in clinical trials. E-stim induced RGC responses were recorded using other electrodes in the outer circumference of the E-stim electrode. The 6-e-hexagon was activated either in All-at-once, Sequential or in Alternating-step activation modes (analogous to L-stim) at different frequencies (1-20 Hz). Activation of E-stim electrodes in Alternating-step outperformed the Sequential activation by an average of 87.78 % increase in RGC responses, suggesting enhanced image perception. Moreover, at a relatively high E-stim frequency of 10 Hz (100 ms step interval) the Alternating-step activation of E-stim electrodes, allowed to maintain RGC responses (average RGC responses = 8.16±0.27 Hz) for the duration of 90 s. In contrast, the All-at-once activation of E-stim electrodes resulted in fading effects within the first few seconds after E-stim (average RGC responses = 0.70±0.19 Hz). Conceptually, the E-stim presentation with topological electrode-activation instead of All-at-once E-stim activation yields reduced fading-effects even at a relatively high E-stim frequency. This outcome was promising for implementation in E-retinal implants. As an example, an image of a Smiley-face was emulated on the electrode array and detected in the respective responses. In conclusion: Implementation of the Alternating-step activation of electrode arrays overcomes the typical image-fading effects and consequently may increase the temporal and spatial resolution of the artificially induced visual perceptions.