Machine Learning for Predicting Diabetes Mellitus Using the Gradient Boosting Machine Algorithm: A Case Study at Al Ihsan Hospital

Diabetes is a persistent ailment that arises when the pancreas fails to generate sufficient insulin or when the body cannot efficiently use the insulin it produces. This study analyzes the application of machine learning with the Gradient Boosting Machine (GBM) algorithm for predicting diabetes mellitus using patient medical records from Al Ihsan hospital. The medical record data used is data for the period 2020 – 2023. The dataset includes attributes such as Gender, Age, Diagnosis, Hemoglobin, Hematocrit, Leukocyte, Erythrocyte, Platelet, HbAlc, and Random Blood Glucose (RBG). The study employs grid search and k-fold cross-validation to determine the optimal GBM parameters. The highest accuracy was observed with the 80:20 split, and the grid search identified the optimal hyperparameters as a learning rate of 0.2, 50 estimators, and a maximum depth of 3. The classification report displays a precision of 0.80, recall of 0.88, and an F1-score of 0.84 for class 0 (DM type 2), and a precision of 0.84, recall of 0.74, and an F1-score of 0.79 for class 1 (DM type 2 with complications). The ROC curve (AUC) for GBM is 0.90, demonstrating its effectiveness in predicting diabetes mellitus types. This study highlights the potential of using advanced machine learning techniques for early detection and management of diabetes mellitus, ultimately contributing to improved patient care and health outcomes.