Automatic Segmentation and Classification of Hard Exudates to Detect Macular Edema in Fundus Images

Detection of the presence of hard exudates in fundus images of diabetic patients is very important. If hard exudates are present in the macular region of retina, it will lead to Diabetic macular edema. Diabetic Macular Edema disease leads to vision loss problem in Diabetic patients. Early detection of Macular Edema in diabetic patients paves a path for prevention from blindness. Diabetic Macular Edema (DME) occurs when blood vessels in the retina of patients with diabetes begin to leak into the macula region of eye. These leakages cause the macula to thicken and swell, progressively leads to vision loss. The automatic detection of Diabetic Macular Edema and classification of DME severity is done in this paper. The Hard Exudates (HE) presence in macula region is detected and the features are extracted. The extracted features are fed as input to Adaptive Neuro Fuzzy Inference System (ANFIS) and Extreme Learning Machine (ELM) classifier to classify the images as normal and abnormal. ANFIS and ELM classifiers performances are evaluated in terms of the parameters such as Sensitivity, Specificity and Accuracy whose values are 100%, 90% and 96.49% for ANFIS classifier and 94.28%, 100% and 96.49% for ELM classifier