A MATHEMATICAL AND SIMULATIVE MODEL FOR ARTIFICIAL INTELLIGENCE-BASED LONG-TERM PREDICTION OF TYPE 2 DIABETES MELLITUS USING MACHINE LEARNING AND DEEP LEARNING MODELS FOR HEALTHCARE ANALYTICS

Abstract

Type 2 Diabetes Mellitus (T2DM) is one of the most prevalent chronic metabolic disorders worldwide and poses a significant challenge to healthcare systems. Early detection and long-term prediction of diabetes risk are essential for reducing complications and improving patient outcomes. This paper presents the development of Artificial Intelligence (AI)-based models for long-term prediction of Type 2 Diabetes Mellitus using the Kaggle PIMA Indian Diabetes dataset. The dataset includes important clinical and demographic features such as glucose level, body mass index (BMI), age, insulin level, blood pressure, pregnancies, and diabetes pedigree function. A systematic methodology was adopted, including data preprocessing, normalization, feature selection, model training, and evaluation. Multiple machine learning algorithms such as Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbours (KNN), along with Artificial Neural Networks (ANN), were implemented and compared. In addition, mathematical modelling and simulative analysis were performed to validate model performance and reliability. The models were evaluated using performance metrics such as accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrix analysis. The results indicate that the Random Forest model achieved the highest prediction accuracy of 93% with balanced precision and recall. Feature importance analysis revealed that glucose, BMI, and age were the most significant predictors of diabetes. The findings demonstrate that AI-based mathematical and simulative models can effectively support early detection and long-term prediction of Type 2 Diabetes Mellitus, contributing to improved healthcare decision-making and preventive strategies.