Detection and grading of diabetic retinopathy from clinical fundus images using machine learning algorithms

Abstract

Diabetes is a disease which affects many important organs of a human body such as eyes heart and so on Around 80% of the patients may have variations in the blood glucose level and may lead to few complications in the eye The patients having diabetes for more than 10 years may be affected by the eye related pathologies such as retinopathy and maculopathy The eyerelated pathologies need appropriate detection diagnosis and treatment to prevent the damage of the eye at the earlier stage The eye related diseases if left untreated can eventually lead to vision loss The retinal disease occurred by the diabetes is Diabetic Retinopathy DR It is a well known problem of visual deficiency and visual impairment among working age grown ups The ophthalmologist identifies DR when the dilated eye exam causes severe in any one of the following deviations in the retina includes changes in blood vessels newly grown blood vessels leaking blood vessels changes in the lens and damages to the nerve tissue and swelling of the macula The earliest detection of DR assists the ophthalmologists to perform the appropriate treatment and thus prevent the patients from vision loss Treatment of DR involved in detecting the symptoms in the form of microaneurysms hemorrhages and exudates in the retinal area The first sign of diabetic retinopathy is microaneurysms and to identify these microaneurysms at initial stage may prevent from DR progression and from vision loss Microaneurysms are saccular capillary dilations which appear as small red dots Retinal hemorrhage is the abnormal bleeding of the blood vessels which occurs in the light sensitive tissue in the retina which is present at the back of the eye newline