Fast and Accurate Prediction of Diabetes Mellitus using Artificial Neural Network

Abstract

Diabetes mellitus, also commonly known as diabetes, is a disease that is spreading rapidly across the world. It is a major metabolic disorder in which the human body is unable to manage its blood glucose levels. Insulin is a hormone secreted by the pancreas that helps convert blood glucose into useful energy for the body. In diabetes, the human body does not produce enough or sufficient insulin and is not able to utilize the insulin it produces properly. This excessive glucose level in the blood causes serious health problems like blindness, heart attack, renal disease, kidney problems, etc., and it leads to death. Patients need to pay a lot of attention to the disease because there is no cure for it and it can be controlled by regular check-ups and medicine. It is necessary to control the growth of diabetes through early detection. So, we are proposing a quick and intelligent system based on an artificial neural network that can help with early detection of diabetes and be helpful to the doctor for early treatment so it can stop the progression of the disease. In this research work, a dataset from Pima, India, was used to predict the possibility of diabetes. The proposed artificial neural network model has an accuracy of 96.35%.