Imaging system for retinal change evaluation

This paper concentrates on the results of a computerised approach to the automatic extraction of numerical indexes describing morphological details of the fundus oculi. The authors proposed an imaging software system with a strict interconnection between the segmentation and recognition phases. This paper presents new image processing techniques developed to take advantage of an improved imaging system constituted by a high resolution digital camera (Kodak DCS 420) connected on top of a standard retinal camera (Topcon TRC-50VT). The higher resolution (1524×1024 pixels) now permits a more accurate analysis of the degree of arteriolar sclerosis and vessel narrowing, both in the proximal and distal segment, and the computation of further numerical indexes, such as vessel reflectance and permeability to fluorescein.