Automatic detection of diabetic retinopathy exudates from non-dilated retinal images using mathematical morphology methods

Diabetic retinopathy is a complication of diabetes that is caused by changes in the blood vessels of the retina. The symptoms can blur or distort the patient’s vision and are a main cause of blindness. Exudates are one of the primary signs of diabetic retinopathy. Detection of exudates by ophthalmologists normally requires pupil dilation using a chemical solution which takes time and affects patients. This paper investigates and proposes a set of optimally adjusted morphological operators to be used for exudate detection on diabetic retinopathy patients’ non-dilated pupil and low-contrast images. These automatically detected exudates are validated by comparing with expert ophthalmologists’ hand-drawn ground-truths. The results are successful and the sensitivity and specificity for our exudate detection is 80% and 99.5%, respectively.