Abstract

Precise and accurate diagnosis of microvascular disease in type 2 diabetic patients is a challenging task due to the asymptomatic behavior of the disease. Machine learning algorithms are used to build classification models for Microvascular Disease classification. Hence this paper filters feature selection techniques for predicting microvascular diseases. Type 2 diabetes is a chronic disease that is a long-term disease and comes with a series of diseases. These diseases are macro and microvascular. Microvascular diseases affect the function of the eyes, and kidneys and damage nerves in the human body. Hence regular health care makes it easy to predict such diseases. These microvascular diseases are predicted using available features in the data. The paper discusses various techniques in feature engineering used by the researcher for predicting microvascular diseases in diabetic patients. Also shows the filter feature selection methods used to improve the accuracy of the prediction. This prediction accuracy is measured through accuracy measures such as precision, recall, and accuracy.