Implementation of Diabetes Incidence Prediction Using a Multilayer Perceptron Neural Network

Diabetes is a long-lasting health condition associated with improper regulation of glucose levels in the body. This chronic disease occurs when blood glucose is too high, causing a variety of complications as well as being the leading cause of death. As the prevalence of diabetes increases, it is critical to plan a new approach for the prevention and management of the disease. However, current disease control methods have limitations in that user convenience and accessibility are low. Recent studies have shown that the development of a disease control system using machine learning is a suitable prediction of diabetes incidences with high accuracy in adults. In this paper, we present a diabetes incidence prediction system using a multilayer perceptron neural network that allows users to enter simple data and increases the convenience and accessibility of the prediction capabilities. The proposed system was evaluated on Pima Indians Diabetes and showed a good performance in predicting diabetes incidences.