IMAGE-PROCESSING STRATEGY FOR THE SEGMENTATION AND QUANTIFICATION OF MICROANEURYSMS IN FLUORESCEIN ANGIOGRAMS OF THE OCULAR FUNDUS

Digital image-processing techniques can provide an objective and highl y repeatable way of quantifying retinal pathology. This study describe s an image-processing strategy which detects and quantifies microaneur ysms present in digitized fluorescein angiograms. After preprocessing stages, a bilinear top-hat transformation and matched filtering are em ployed to provide an initial segmentation of the images. Thresholding this processed image results in a binary image containing candidate mi croaneurysms. A novel region-growing algorithm fully delineates each m arked object and subsequent analysis of the size, shape, and energy ch aracteristics of each candidate results in the final segmentation of m icroaneurysms. The technique is assessed by comparing the computer's r esults with microaneurysm counts carried out by five clinicians, using Receiver Operating Characteristic (ROC) curves. The performance of th e automated technique matched that of the clinicians' analyses. This s trategy is valuable in providing a way of accurately monitoring the pr ogression of diabetic retinopathy. (C) 1996 Academic Press, Inc.