Purpose:
Electrode placement is a major issue in the design and surgical placement of epiretinal prostheses. For proper stimulation of the surviving retinal ganglion cells, it is desirable to maximize tissue-device contact area. The aim of this study was to examine the variation in slope along the retinas of 13 retinitis pigmentosa (RP) subjects.
Methods:
Retinal contour images were acquired using spectral domain optical coherence tomography (Heidelberg Spectralis OCT) in both eyes of 13 subjects with RP. Surface contour lines were drawn using Heidelberg Eye Viewer Software, while Igor Pro was used to generate point clouds that describe thickness as a function of horizontal scan position. Next, Matlab was used to calculate first and second derivatives of the curvature values, yielding the slope gradients and inflection points along the retina (see Figure). Finally, a simple Microsoft Excel algorithm was used to determine the regions of the retina that display the smallest changes in slope, as well as inter-patient variability.
Results:
Plots were generated to display the slope profiles along the retina. The attached figure is one example of a retinal profile generated using Matlab for one patient. Using the left eye of 13 subjects with RP, the regions of the retina having second derivative values less than or equal to 0.05 were found to be 1.71±0.62 to 2.06±0.54 mm temporal to the foveola and 1.82±0.48 to 2.11±0.46 mm nasal to the foveola. Using the same eye and patients, the regions of the retina having second derivative values less than or equal to 1 were found to be 1.50±0.43 to 2.50±0.34 mm temporal to the foveola, and 1.48±1.16 to 2.15±1.55 mm nasal to the foveola. A slope differential of 0.05 corresponds to an angle change of 0.3o while a slope differential of 1 corresponds to an angle change of 4.45o.
Conclusions:
Plotting methods using Igor Pro and Matlab, as well as analysis in Excel, provide a reasonable approach to analyze retinal curvature and estimate flatness in RP patients. The slope profiles may aid in the estimation of the contact surface area between a retinal prosthesis implant and the retinal tissue; the regions of the retina having smallest second derivative values will provide the area of maximum contact between the device and the tissue.
Keywords: imaging/image analysis: clinical • retina • retinitis