Smart Solutions for Continuous Blood Glucose Level Detection

Abstract

Diabetes is a prevalent chronic illness affecting millions of individuals worldwide, necessitating vigilant monitoring and management of blood glucose levels. In this study, the primary objective is to predict patients' blood glucose levels using Machine Learning algorithms. Various parameters, including blood pressure, sex, diabetes pedigree function, BMI, age, insulin, and skin thickness, are considered as input features for the predictive models. Four machine learning algorithms, namely Logistic Regression, Support Vector Machine, Artificial Neural Networks, and Deep Learning Neural Network, are employed to forecast the likelihood of diabetes in patients. The process involves gathering patient data and feeding it into the selected algorithms, which then generate predictions for the blood glucose level. By comparing the outputs produced by the four algorithms, the most effective model for predicting diabetes is determined. Accurate predictions can significantly aid in early detection and proactive management of diabetes, leading to improved patient outcomes and better quality of life. This research contributes to enhancing diabetes care by leveraging the potential of Machine Learning in predicting blood glucose levels and supporting healthcare professionals in making informed decisions for effective diabetes management.