A Literature Review On Decision Support System Models To Predict Diabetic Retinopathy Using Historical Data, And Machine Learning Techniques

Abstract

The expansion of health sciences and IT has been made possible by electronic health records, which in turn have led to the creation of enormous data sets. Knowledge Discovery from Data (KDD) or data mining refers to the technique of automatically finding patterns that stand in for knowledge. Massive databases, data warehouses, or information repositories hold all the known information. Damage to the retina's blood vessels is known as diabetic retinopathy. The retina is the rear of the eye's light-sensitive layer of cells. The results of this are a reduction in blood flow and blurred eyesight. In addition, swelling and hemorrhage are seen. New blood vessels may start to sprout at some stage, but further problems may develop. Most cases of diabetic retinopathy affect both eyes. Diabetes mellitus type 2, gestational diabetes, and insulin-dependent diabetes can all lead to diabetic retinopathy. Those who have lived with diabetes for an extended period are more likely to develop retinopathy. It is critical to develop a Clinical Decision Support System to assist physicians in establishing the diagnosis of this condition because of how challenging it is. Knowledge extraction from medical databases is crucial for accurate medical condition diagnosis. This research aims to conduct a comprehensive literature evaluation of machine learning approaches used by decision support systems and clinical decision support systems for data balance and diabetic retinopathy prediction.