The Incidence of Diabetes Prediction Based on the Logistic Regression Model

Abstract

Diabetes is a rapidly growing global public health challenge that demands effective prevention and treatment strategies. This study aims to explore the relationship between diabetes incidence and relevant indicators such as BMI, blood pressure, and skin thickness by utilizing data from the Kaggle dataset. In this study, the logistic regression model was employed to identify risk factors associated with the incidence of diabetes. The logistic regression model allows this study to test the effect of correlation between predictor variables and the outcome variable, and develop a model to predict the likelihood of diabetes incidence. Using a Python implementation of the model, nearly 77% precision in predicting the incidence of diabetes was obtained. The analysis revealed a strong correlation between age and BMI and the incidence of diabetes, aligning with previous findings in the domain knowledge of diabetes. The results of the study may help individuals and healthcare providers to identify and manage the risk factors associated with diabetes, ultimately reducing the incidence of this disease. The proposed approach of utilizing the logistic regression model provides a valuable tool for predicting the onset of diabetes, contributing to the ongoing effort to combat this global public health issue.