Figure 1

Representation of the algorithm used to investigate the images. The en face RPE elevation map image was used to identify regions occupied by drusen. This image was imported into ImageJ, and the “MaxEntropy” threshold was applied to “binarize” this image. In the binarized image, the “drusen” region (red) was excluded in a following analysis which was aimed at measuring metrics in the “drusen-free” region. Measurements were performed in a circular region of interest centered on the fovea (diameter of 5.0 mm—displayed in yellow in the figure). The en face image of the ellipsoid zone (EZ) was also imported in ImageJ. A previously described image-processing algorithm using two reference structures, the vitreous and the retinal nerve fiber layer, was employed to calculate the EZ normalized reflectivity. The en face flow image of the CC was compensated using the corresponding CC en face structural image. A local thresholding method was applied in order to detect pixels falling below this threshold (in red in the lower right image). Values detected in this example are, as follows: drusen area = 1.36 mm²; drusen volume = 0.05 mm³; normalized EZ reflectivity in the ROI = 0.79; normalized EZ reflectivity in the drusen-free region = 0.80; CC FD% = 24.2%; CC FD% in the drusen-free region = 24.1%.