A Deep Meta-learning Framework for Heart Disease Prediction

Heart disease is a leading cause of death. This is proved by the high mortality rates annually published. The factors and symptoms of this disease which are significantly common among other diseases, grabbed the attention of the scientific community to integrate advanced tools and computing systems such as Machine Learning (ML), and Artificial Neural Networks (ANN) to early diagnose and predict this disease. Among recent research work, Restricted Boltzmann Machine (RBM) formed a viable model to solve similar problems in the medical domain. Therefore, we introduce this paper to present a heart disease classification system based on RBM. In this paper, we relied on a standard dataset, namely the Cleveland dataset to conduct our experimental work. In addition, we developed a meta-learning framework to train stacked RBM classifiers, where we use the same type of classifier with different levels of abstractions of the data to get several classifications. The experimental results show that our proposed model achieves an accuracy that is on par with the state-of-the-art.