Building Classification System to Predict Risk factors of Diabetic Retinopathy Using Text mining

This Making medical decisions such as diagnosing the diseases that cause a patient's illness is often a complex task. The Diabetic retinopathy is one of the complications of diabetes and Diabetic retinopathy is one of the most common causes of blindness. Unfortunately, in many cases the patient is not aware of any symptoms until it is too late for effective treatment. Analysis of the evoked potential response of the retina, optical nerve and optical brain centre will pave a way for early diagnosis of diabetic retinopathy and prognosis during the treatment process. The objective of this study is to identify the prevalence and severity of diabetic retinopathy and to determine the relationship between risk factors, prevalence and severity of diabetic retinopathy. We collected 3450 patients history, who are suffering with type 2 diabetes .As the available data is not in structured format, we apply text mining classification technique to predict the risk factors of the diabetic retinopathy. This study shows that a relatively short duration of case management instituted before onset of clinically identifiable retinopathy, significantly reduce the risk of developing retinopathy in patients with type 2 diabetes. A total of 1402 patients (39.8%) had evidence of retinopathy. This comprised of 32% of initial stage of DR , 20% Retinal haemorrhages, 14% patients with Mild non proliferate diabetic retinopathy, 18% with Moderate non proliferate DR, 1% with Proliferate DR ,14%with High risk.