Early Detection of Coronary Heart Disease Based on Machine Learning Methods

Abstract

Heart disease detection using machine learning methods has been an outstanding research topic as heart diseases continue to be a burden on healthcare systems around the world. Therefore, in this study, the performances of machine learning methods for predictive classification of coronary heart disease were compared. Material and Method: In the study, three different models were created with Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) algorithms for the classification of coronary heart disease. For hyper parameter optimization, 3-repeats 10-fold repeated cross validation method was used. The performance of the models was evaluated based on Accuracy, F1 Score, Specificity, Sensitivity, Positive Predictive Value, Negative Predictive Value, and Confusion Matrix (Classification matrix). Results: RF 0.929, SVM 0.897 and LR 0.861 classified coronary heart disease with accuracy. Specificity, Sensitivity, F1-score, Negative predictive and Positive predictive values of the RF model were calculated as 0.929, 0.928, 0.928, 0.929 and 0.928, respectively. The Sensitivity value of the SVM model was higher compared to the RF. Conclusion:Considering the accurate classification rates of Coronary Heart disease, the RF model outperformed the SVM and LR models. Also, the RF model had the highest sensitivity value. We think that this result, which has a high sensitivity criterion in order to minimize overlooked heart patients, is clinically very important.

Prediction of Heart Disease Using Machine Learning Algorithms

“…When comparing the three classification models, we can see that the accuracy of Random Forest algorithm is 95.63% which is higher than the accuracy of the other two classification techniques. In addition, if we have contrasted the results with past research, as was done in part IV of the problem statement portion, we attained the best accuracy using Random Forest, which is greater than [1] and is 95.63%. Yilmaz et al [1] obtained the best accuracy from Random Forest, which is 92.90%.…”

“…Due to their objectives are comparable to ours, we used Yilmaz et al [1], Pal et al [2], and Rajdhan et al [8] as our foundation articles in this study. The authors [1], [2] and [8] have used all the attributes present in the dataset and achieve good accuracy from their classification models but they have used small size dataset to compare classification model accuracy as well as did not handle null values present in the dataset. In addition, they did not employ any method of feature selection to identify strongly correlated features that can enhance classification model accuracy.…”

“…The identification of heart disease and many other major diseases has been the subject of several studies utilizing various data mining techniques. In order to predict coronary heart disease, Yilmaz et al [1] used a variety of machine learning techniques. In comparison to other models, Random Forest (RF) has provided the greatest accuracy, with a score of 92.90%.…”

Artificial Intelligence-based Colon Cancer Prediction by Identifying Genomic Biomarkers

“…For classification, trees each leaf node is created to contain only members of one class. For regression, trees continue to divide until a small number of units remain in the leaf node (12).…”

Classification of colorectal cancer based on gene sequencing data with XGBoost model: An application of public health informatics

“…ML methods are one of the technologies that have seen widespread use in disease diagnosis and clinical decision support systems in recent years, and they have a wide range of applications. ML methods are typically used to classify disease prediction 14,15 . ML, which has a wide range of applications in the field of health, is the foundation of applications in the determination of genetic diseases, early detection of cancer diseases, and pattern recognition in medical imaging 16 .…”