Machine Learning Predictive Models for Coronary Artery Disease

Abstract

Coronary artery disease (CAD) is the commonest type of heart disease and over 80% of the deaths resulted from the diseases occurred in developing countries including Nigeria, with majority being in those victims are below 70 years of age. Though, CAD is not a well known disease in Nigeria but however in year 2014, 2.82% of the total of deaths occurred in the country were due to the disease. In this study, a machine leaning predictive models for CAD has been developed with diagnostic CAD dataset obtained in the two General Hospitals in Kano State—Nigeria. The dataset applied on machine learning algorithms which include support vector machine, K nearest neighbor, random tree, Naïve Bayes, gradient boosting and logistic regression algorithms to build the predictive models and the models were evaluated based accuracy, specificity, sensitivity and receiver operating curve (ROC) performance evaluation techniques. In terms of accuracy random forest-based machine learning model emerged to be the best model with 92.04%, for specificity Naive Bayes based machine learning model emerged to be the best model with 92.40%, while for sensitivity support vector machine based machine learning model emerged to be the best model with 87.34% and for ROC, random forest-based machine learning model emerged to be the best model with 92.20%. The decision tree generated with random forest machine learning algorithm which happened to be best model in terms accuracy and ROC can be converted into production rules and be used develop expert system for diagnosis of CAD patients in Nigeria.

Multivariate sample similarity measure for feature selection with a resemblance model

“…Table 3. Comparison of the existing and proposed feature selection approach Method Dataset Algorithm employed Accuracy achieved (%) [7] Heart disease XGBoost 90 [8] Heart disease Naïve Bayes 85 [9] Heart disease LR 90.6 [10] Heart disease SVM 90.65 [11] Heart disease k-NN 90 [12] Heart disease SVM 98.7 [13] Heart disease RF 92.04 [14] Heart disease GNB 94.92 [15] Heart disease SVM 95 Proposed method Heart disease XGBoost 99.32…”

“…Wankhede et al [13] and Muhammad et al [14] used feature selection on a heart disease dataset to develop a linear support vector machine (SVM) model for heart disease diagnosis. The developed SVM model has an area under curve (AUC) score of 0.96 and a precision of 90.65%.…”

Long-Term Coronary Artery Disease Risk Prediction with Machine Learning Models

“…In [ 68 ], the authors applied ML algorithms, including SVM, KNN, RT, RF, NB, gradient boosting (GB) and LR, on a dataset obtained in the two General Hospitals in Kano State, Nigeria for the prediction of CAD. In terms of accuracy, the random forest model emerged as the best model with 92.04%; for specificity, the NB model was the best, with 92.40%.…”

Current and Future Applications of Artificial Intelligence in Coronary Artery Disease

“…Therefore, improving the ability to predict CAD using more accurate risk-assessment modeling is imperative, given the potential to reduce downstream testing and associated costs. Using clinical and demographic features, ML models have been employed to estimate the PTP of CAD [ 32 , 33 , 34 ]. In a recent multicenter cross-sectional study, a deep neural network algorithm based on the facial profile of individuals was able to achieve a higher performance than traditional risk scores in predicting PTP of CAD (AUC for the ML model 0.730 vs. 0.623 for Diamond –Forrester and 0.652 for the CAD consortium, p < 0.001) [ 35 ].…”