NOVEL EARLY DIAB EDI DEVICE FOR PREDICTING TYPE 2 DIABETES

Abstract

The rising prevalence of Type 2 diabetes mellitus (T2DM) poses significant public health challenges globally, necessitating early detection and intervention strategies to mitigate its impact. This study investigates the application of machine learning (ML) algorithms for the prediction of T2DM, utilizing a comprehensive dataset that includes demographic, clinical, and lifestyle factors. Various ML models, including logistic regression, decision trees, random forests, and support vector machines, were employed to identify key predictors and enhance the accuracy of diabetes prediction. The performance of these models was evaluated using metrics such as accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC-ROC). Our findings demonstrate that the random forest model outperformed other algorithms, achieving an accuracy of 88.5% and an AUC-ROC score of 0.92, indicating its robustness in predicting T2DM. The logistic regression model followed with an accuracy of 84%, while decision trees and support vector machines achieved accuracies of 81% and 79%, respectively. Additionally, feature importance analysis revealed that factors such as body mass index (BMI), age, and family history significantly influenced the risk of developing diabetes. The results underscore the potential of ML techniques as effective tools for early diabetes prediction, facilitating timely intervention and personalized treatment strategies. This study contributes to the growing body of literature advocating for the integration of machine learning in diabetes management and encourages future research to explore more complex models and larger datasets for improved predictive performance.