An Optimal Microaneurysm Detection Technique in Digital Fundus Images for DR Recognition

In this method, an optimal combination of the internal components of Microaneurysm detectors is used to detect microaneurysms and Diabetic Retinopathy grading is done based on the number of microaneurysms present. Diabetic Retinopathy is a serious eye disease which is caused due to Diabetes. The means of early recognition of diabetic retinopathy is to detect microaneurysms in the retinal fundus on time. Microaneurysms are small dark red spots that appear on the surface of the retina. In medical image processing, reliable detection of microaneurysms is still an open issue. In this method, several preprocessing methods and candidate extractors, which are the internal components of microaneurysm detectors are combined. This system ensures high flexibility by using a modular model and a simulated annealing-based search algorithm is used to find the optimal combination. Images from Messidor database which is publicly available is used for testing. DR grading is done based on the presence or absence of microaneurysms and the number of microaneurysms present.