article · Scientific Reports
Managing diabetes mellitus effectively requires early and precise identification to avoid severe health complications. A machine learning framework using Extreme Gradient Boosting, or XGBoost, categorises patient health into non-diabetic, pre-diabetic, and diabetic states. Evaluated on a cleaned clinical dataset containing 826 unique records, the pipeline applied rigorous data preparation, including a stratified train-test split and five-fold cross-validation. The standard XGBoost model delivered an accuracy of 99.60 percent, maintaining an average cross-validation accuracy of 98.79 percent. Balancing techniques such as Random Over Sampling and Synthetic Minority Over-sampling Technique matched this top accuracy while enhancing generalisation, albeit requiring greater computational resources. Conversely, under-sampling strategies degraded accuracy due to lost information. Comparative benchmarking confirmed that the gradient boosting approach outperformed alternative algorithms, including Random Forest, Logistic Regression, and Support Vector Machines.
Accurate multi-class prediction allows healthcare providers to identify individuals in pre-diabetic stages before severe metabolic damage occurs. By reliably separating pre-diabetic cases from non-diabetic and diabetic conditions, this machine learning method provides a computational foundation for detecting metabolic risk early, enabling timely medical interventions and better long-term disease management.
The framework could inform the development of clinical decision-support software for healthcare practitioners screening patients for diabetes risk. Because the evaluation relies on a retrospective dataset of 826 clinical records, the underlying technology represents early-stage algorithm development. Transitioning toward real-world clinical use would require validating the pipeline across larger, multi-centre datasets and embedding the predictive models into diagnostic workflows.
AI-generated from the published abstract. Always read the original work before citing.
Diabetes mellitus is a persistent metabolic condition that requires accurate and early diagnosis to prevent severe complications. This paper proposes an Extreme Gradient Boosting (XGBoost)-based predictive framework for multi-classification of diabetes mellitus into non-diabetic, pre-diabetic, and diabetic classes. After standardization and exclusion of non-clinical identifiers, Duplicate clinical records were removed from the original dataset, leaving 826 unique records. Comprehensive preprocessing pipeline used a stratified 70:30 train-test split and five-fold cross-validation; scaling and resampling were performed only within training partitions. Experimental results on the original dataset XGBoost achieved an accuracy of 99.60%. Both Random Over Sampling (ROS) and Syntenic minority over sampling technique (SMOTE) have also achieved 99.60% accuracy but provided improved generalization at the expense of higher computational cost. In contrast, Random Under Sampling (RUS) and Cluster Centroids (CC) reduced accuracy to 92.74% and 90.32%, respectively due to information loss. Across five folds, the original XGBoost model achieved 98.79 ± 1.13% accuracy. Benchmarking against Logistic Regression, Random Forest, Support Vector Machine, Decision Tree, and K-Nearest Neighbors showed that XGBoost provided the strongest performance. These findings highlight the effectiveness of XGBoost while emphasizing classification accuracy in multiclass diabetes prediction systems.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1038/s41598-026-65844-3
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
