Earmarking Retinal Changes in a Sequence of Digital Color Fundus Photographs

This paper presents a novel approach for earmarking retinal changes in a sequence of digital color fundus photographs, using Principal Component Analysis (PCA) and a temporal color-coded scheme so as to improve the representation of the global effect of these changes — its aim is to assist graders in detecting alterations on digital color photographic follow-up time-sequences of the human retina. The color-coded scheme was developed so that: the earliest earmarking color of areas which have been detected in multiple instances in the followup timeline is kept; areas which have suffered changes but have returned to their original appearance are earmarked with a specific color-code. The composite image formed by the projection of the color-code image onto the baseline image used as reference then guides the grader’s attention to changes detected on the fundus photographs.