Improving Coronary Heart Disease Prediction Through Machine Learning and an Innovative Data Augmentation Technique

Coronary heart disease (CHD) is a leading cause of death globally, with over 382,000 deaths in the USA alone in 2020. The early detection of CHD is critical in reducing mortality rates. Artificial intelligence (AI) is a constantly evolving field of computer science that employs computational models to extract insights from past data and provide rapid and accurate predictions for future cases. This paper presents a novel approach that generates an augmented dataset by selectively duplicating misclassified instances during the leave-one-out cross-validation (CV) process to overfit a model. We used a paired machine learning model with an augmented dataset approach to evaluate several classifiers. The comprehensive heart disease dataset [1] served as our base dataset. Our approach achieved higher accuracy than the base dataset, with the bagged decision tree (DT) algorithm outperforming state-of-the-art models and achieving an accuracy of 97.1% in the 10-fold CV test. Further experiments using the Cleveland dataset and the same 10-fold CV test resulted in an even higher accuracy of 99.2%. Combining an augmented dataset and the bagged-DT algorithm holds great promise for early CHD prediction helping reduce CHD mortality rates. The use of AI in early CHD prediction could potentially make a difference between the life and death of the patient.

Automated summary

Coronary heart disease (CHD) is a major cause of death globally, with over 382,000 deaths in the USA alone in 2020. Early detection is crucial for reducing mortality rates. This paper proposes a novel approach that uses an augmented dataset and a machine learning model to achieve higher accuracy in CHD prediction. The bagged decision tree algorithm outperforms other models, with an accuracy of 97.1% in the 10-fold cross-validation test.