Abstract:
Purpose: Retinal prosthetic devices have been trialled in patients and encouraging results have been reported. Various factors including retinal disease state and effective electrode placement during surgery, would affect the outcome of retinal prosthesis implementation. In this study, we explore the possible effect of implant foveal eccentricity on retinal activation threshold using computational modelling. Method: Five healthy and five early stage dry age-related macular degeneration human volunteers were scanned using optical coherence tomography. The 3D retinal structure of each individual was extracted from the images and smoothly meshed. A set of five hexapolar electrode configurations was placed in the sub-foveal, 1 mm and 2 mm eccentricity positions from the centre of the retina within the subretinal space. The activation threshold for individual electrodes at all three sites of stimulation were simulated, using bidomain finite element modelling, and compared for all subjects. Results: We found that sub-foveal pit electrode placement required the lowest activation threshold, followed by the 1 mm eccentricity electrode placement. The 2 mm eccentricity electrode placement required the highest activation threshold. Conclusion: Our simulations indicate that clinically minute changes in foveal eccentricity of the subretinal implant can increase activation threshold and therefore, with a given stimulus current, retinal prosthetic performance drops with foveal eccentricity. Translational Relevance: Retinal prosthetic devices have been introduced as a potential therapy for patients suffering from outer retinal diseases. Human subjects with retinal implants demonstrated a varying range of responses to these devices. In this study, we introduce an anatomically-based patient-specific model of the human retina, using it to explore the effect of foveal eccentricity on activation threshold.