Using Machine Learning for early prediction of Heart Disease

Heart disease is today the leading cause of death in the world. The early diagnosis can significantly reduce the risk of death and can be useful to drive successful treatment. However, the early diagnosis requires continuous monitoring of a large set of clinical and lifestyle indicators. This is the reason why there is an increasing number of studies aimed to adopt machine learning to predict heart disease starting from the analysis of the vast range of clinical data that we can collect today thanks to the advent of patients' digital folders. This work investigates the adoption of a large set of machine learning and deep learning classifiers to predict heart disease from the data gathered by a proposed feature model. The study also proposes hyperparameters optimizations aimed to improve the performance of the adopted classifiers. The evaluation is performed on a real dataset and the obtained results show good performance.