oa Preliminary studies for thalamic visual prosthesis to treat blindness resulting from glaucoma: A deep brain stimulation approach

Abstract- Developing visual prosthesis in human cortex or on retina has gained some importance over years. However, visual prosthesis in thalamic region targeting lateral geniculate nucleus (LGN) has few advantages though. This paper proposes a simulation model for deep brain visual prosthesis that will harness the capacity of the residual central visual neurons. This prosthesis creates percepts, substituting an engineered prosthetic for the image capture and early visual processing properties of the LGN neurons that we plan to drive artificially. Introduction- Glaucoma is an incurable neurodegenerative disorder leading to permanent blindness. Glaucoma accounts for blindness in 4-5 million of the world's people and is unusually prevalent in Eastern Arabia. In Qatar, glaucoma is said to account for 40% of its cases of blindness and 16% of people above age 40 are affected by this disease. Restoration of visual function in glaucoma by stem cell therapy or genetic engineering could provide a quite good solution, but the likelihood of these approaches providing a cure for glaucoma in the near term is tenuous. A neuroprosthetic approach offers the most likely short-term treatment for glaucoma-induced blindness. This paper describes early progress made towards designing a visual prosthesis to treat blindness through glaucoma. In glaucoma, there is evidence that the LGN undergoes degenerative changes in human patients [1]. Obviously, LGN degeneration during glaucoma needs to be factored into the design of the electrode array. Design and Concept- The proposed design of the visual prosthetic requires many components (Fig 1). The electrodes are positioned in the dorsal layers of the LGN and all other components located outside the body. Images are captured and adjusted for position of gaze and processed by spatiotemporal filters to provide signals that can be used to drive LGN neurons similar to their natural drive. These signals drive the electrode array with a pattern of stimulation pulses. Electrical stimulation circuitry provides charge-balanced current pulses to the electrodes to activate LGN neurons. Results- An array of electrodes distributed across a semi-elliptical shaped flat surface that interfaces with the top three LGN layers (layers 6-4) is designed. The flat semi-elliptical shaped head, resembling the tip of a knife, facilitates easy penetration through the soft brain tissue. To match LGN dimensions, the head would be 2.5 mm wide, 2 mm deep and 50 µm thick. Fig 2 shows the proposed design with 184 hemispherical shape electrodes placed on either side of the lead with diameters of 50 µm and center to center spacing of 200 µm. The hemispherical shape of electrodes avoids concentration of current density at one place since it does not have any corners or edges. Thus these electrodes deliver uniform current density around the electrode minimizing the electrode and tissue damage. The electrodes are placed in a triangular array to maximize packing density and maintain equal distance between neighboring electrodes. Reference- [1] Gupta, N. et al., 2009. Atrophy of the lateral geniculate nucleus in human glaucoma detected by magnetic resonance imaging. The British journal of ophthalmology, 93(1), pp.56-60.