Diabetic Retinopathy (DR) is the most serious complication eye disease of diabetics’ patients, it occurs when the small blood vessels have a high level of glucose causing a change in the retina, which occurs over a period of time under diabetics, this change cause blur vision and if left undiagnosed and untreated it can eventually lead to blindness. Exudates are one of the primary signs of DR, they appear as yellowish areas with varying sizes, shapes, and locations about areas of leakage, therefore early detection and timely treatment can prevent and delay the risk of vision loss.

Current methods of DR detection are manual, expensive, and require trained ophthalmologists, so it was therefore thought to find an alternative method. Automatic computerized screening could facilitate the screening process, reduce inspection time, and increase accuracy which is vital in ophthalmic treatment.

In this research, we proposed an automatic method to detect exudates from retinal digital images with non-dilated pupils of retinopathy patients based on fuzzy c-means (FCM) clustering technique with a combination of morphology and pre-processing techniques.

Before detecting the exudates, we eliminate both OD and blood vessels network from the retinal image, a preprocessing of contrast enhancement is applied to enhance the quality of the input image. Afterwards, the most effective features are extracted, then used as input data for FCM method.

Finally, the detection overall performance is evaluated by comparing the successful detected exudates with the ground truth (GT), that are drawn from our expert ophthalmologist, by measuring sensitivity, specificity and accuracy, which found to be very promising on the testing studied database

عنوان المؤتمر

المؤتمر الفلسطيني السادس للتوجهات الحديثة في الرياضيات والفيزياء

دولة المؤتمر

فلسطين

تاريخ المؤتمر

5 أغسطس، 2018 - 7 أغسطس، 2018

راعي المؤتمر

جامعة خضوري