What is the clinical evidence from this study?

Study design: Other. Population: Diabetes (n=520). Intervention: Random Forest machine learning model vs. Other machine learning algorithms. Primary outcome: Mean prediction accuracy.

January 11, 2024Open Access

Machine learning algorithms-based decision support model for diabetes

Key Result

The Random Forest machine learning model achieved the highest mean accuracy of 98.13% and an AUC of 1.0 in predicting early-stage diabetes risk using 10 selected clinical features.

Structured PICO

Population

520 patients from Sylhet Diabetes Hospital in Bangladesh whose clinical features were used to develop and validate machine learning models for early-stage diabetes risk prediction.

Intervention

Machine learning algorithms (Random Forest, XGBoost, LightGBM) utilizing Boruta feature selection

Outcome

Model accuracy and Area Under the ROC Curve (AUC) for diabetes risk prediction

A Random Forest machine learning model utilizing Boruta feature selection achieved 98.13% accuracy in predicting early-stage diabetes risk.

Limitations

Relatively small dataset size of 520 instances
Requires future validation on larger datasets using deep learning approaches

Abstract

This research explores the application of machine learning (ML)-based risk prediction models in early diabetes disease detection for healthcare professionals. Diabetes affects millions of people worldwide. In light of significant advancements in biomedical sciences, vast volumes of data have been generated, including high-throughput genetic and diagnostic data sourced from extensive health records. Leveraging an initial diabetes risk prediction dataset from the University of California Irvine (UCI) ML repository, our research focused on supervised learning techniques, constituting 85% of the employed methods. The remaining 15% comprised unsupervised learning approaches, specifically association rules. A key contribution of this study lies in the development of an optimal prediction model utilizing supervised ML algorithms. The Boruta feature selection algorithm was employed to identify pertinent features, and the subsequent models were validated using a preprocessed dataset containing 10 attributes. Notably, the risk prediction models generated through random forest, extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM) exhibited impressive average accuracies of 98.13%, 97.37%, and 97.22%, respectively, as determined via 10-fold cross-validation with 15 repetitions. Furthermore, these models achieved exceptional area under the ROC curve (AUC) values of 1, 0.99, and 0.99, respectively, showcasing their robustness and efficacy in diabetes risk prediction.

Ask AI

Helpful

Bookmark

View Full Paper