FULLY AUTOMATED DIABETIC RETINOPATHY MONITORING AND GRADING SYSTEM

Diabetic retinopathy (DR) is an eye-related disease due to diabetes mellitus. Mainly, the complication is cause by losses of capillaries in the perifoveal capillary network inside retina. Retina Fundus image, Foveal avascular zone (FAZ) that located near macula can be used for diagnosis of DR. Research show that the radius of FAZ can determine the severity level of DR. Semi-automated DR grading and monitoring system has been developed for clinical practice. This system had been proven to achieve high sensitivity, accuracy and specificity. However the process need manual selection of center of FAZ which can be uncertained to the user other than train operator or ophthalmologists. The objective of this project is to eliminate the user input by integrating an automatic selection center of macula algorithm with existing semi-automated DR grading and monitoring system. This is to make DR grading and monitoring system fully automated. In automatic selection center of macula algorithm, optic disc which is landmark of retina fundus image will be identified first.