Voxel data from a three-dimensional optical coherence tomography (3-D OCT) scan of a retina and pixel data from a two-dimensional (2-D) fundus image are spatially mapped. A 2-D composite image generated from the 3-D OCT data is spatially mapped to a fundus image using spatial indicia common to both images. The 3-D OCT data is then spatially mapped to the fundus image. An image processing system generates cross-correlated graphical representations of 3-D OCT data, subsets of 3-D OCT data, and a fundus image.