Forecasting Efficient Prediction of Diabetic Coronary Heart Disease Using Hybrid Ensemble Learning Method (HELM) and Feature Selection Algorithm

Abstract

Artificial intelligence is a component of machine learning, which is utilized in data science to solve several issues. Predicting a result based on available data is a popular use of machine learning. Diabetic Coronary Heart Disease (DCHD) poses a significant healthcare challenge globally, demanding accurate predictive models for early detection and intervention. This study proposes a novel approach that integrates hybrid ensemble learning techniques with feature selection algorithms to enhance efficiency and accuracy of DCHD prediction. The hybrid ensemble learning method combines the multiple ensemble classifiers and compare with machine learning (ML)algorithms like decision trees (DT), support vector machine (SVM), and neural networks (NN), adaboost, XGboost to form a robust predictive model. Additionally, feature selection algorithms are employed to identify the most relevant predictors from a comprehensive set of clinical and demographic variables associated with DCHD. The proposed framework is validated using a large dataset comprising demographic information, medical history, and laboratory test results of diabetic patients. Performance evaluation metrics, including accuracy, sensitivity, specificity, and ROC curve, are utilized to assess the predictive performance of the model. Results demonstrate that the hybrid ensemble learning approach, coupled with feature selection algorithms, outperforms traditional single-model methods in terms of predictive accuracy and efficiency. The selected features provide valuable insights into the underlying risk factors and biomarkers associated with DCHD, facilitating early detection and personalized intervention strategies. Overall, this research contributes to the advancement of predictive modeling in healthcare by offering a sophisticated yet interpretable approach for forecasting DCHD risk in diabetic patients. The proposed framework has the potential to improve clinical decision-making and resource allocation, ultimately leading to better management and prevention of diabetic complications, including coronary heart disease.