FUZZY INFERENCE SYSTEM AIDED DAMAGE ASSESSMENT OF DIABETIC MACULOPATHY USING RETINAL IMAGES

In this paper we elaborate the development of a Fuzzy Inference System (FIS) model for damage assessment of Diabetic Maculopathy. The system is developed using 100 images availed from public databases and also from ophthalmic clinic. The development of the model consists of: i)Pre-processing of the images ii) delineation of ROI iii) extraction of features such as optic disk diameter, number of spots, area of major spread of exudates, major axis length & minor axis length of largest exudate and the degree of yellowness and iv) development of FIS and finally damage assessment in terms of low, medium, high and very high severity. The evaluation of the system was done using metrics such as sensitivity, specificity and accuracy. These metrics are in the range of 94% to 100% heralding efficiency of FIS model.