Categorization of EOG signal for Diabetic retinopathy detection

A Consequence of diabetes is diabetic retinopathy, which is brought on by damage to the blood vessels in the retina, a tissue that is sensitive to light. It is the primary cause of blindness and visual impairment in people of working age, and it is the most frequent cause of vision loss in diabetics. The electrooculogram is taken for 30 subjects with diabetic retinopathy and compared with normal subjects. The difference in the pattern of the waveform of Electrooculogram (EOG) helps in differentiating the disease. Both normal and diabetic retinopathy individuals' filtered signals are subjected to mathematical characteristics and eight important features are taken for classification with the help of feature scores. K-means clustering helps in the clear categorization of data. In this paper, we also proposed cross-validation scores of 3-fold, 5-fold, and 10-fold Random Forest classifiers. With a 90% accuracy rate in automatically detecting DR, this method may help make a very reliable referral recommendation for additional testing and treatment.