One of the essential parts of human life is healthcare. Heart disease is one of the worst diseases, claiming the lives of millions of people all over the world. Accordingly, heart disease prediction is considered a significant aspect of clinical data analysis. Therefore, numerous research has been conducted to create machine-learning algorithms for the early detection of heart disorders in order to assist clinicians in the design of medical procedures. Traditional methods have been limited by their inability to generalize adequately to new data not seen in the training set. This paper proposes a hybrid bidirectional LSTM and 1D CNN architecture with Bayesian optimization for hyperparameters to increase the accuracy of heart disease prediction. The performance of the proposed 1D CNN-BiLSTM approach is validated via evaluation metrics, namely, accuracy, specificity, sensitivity, and area under the receiver operating characteristic (ROC) curve by using two datasets from the University of California, namely, Cleveland and Statlog. The experimental results confirm that the proposed approach attained the high heart-disease-prediction accuracies of 89.01% and 82.72% on the Cleveland and Statlog datasets, respectively. Furthermore, the proposed approach outperformed other state-of-the-art prediction methods.
|
