Heart Stroke Risk Analysis: A Deep Learning Approach

Heart Stroke is one of the severe health hazards; therefore, early heart stroke prediction helps the society to save human lives. This objective can be achieved using the machine learning techniques. In this research article, machine learning models are applied on well known heart stroke classification data-set. In addition, effect of pre-processing the data has also been summarized. In the experimental analysis, machine learning models and ANN with standard feature selection technique are tested on data-set, framingham and the obtained results are evaluated using the confusion metrics includes recall, F1-score, precision and accuracy. From the obtained results, it is observed that ANN performed the best on the pre-processed data, giving the highest accuracy of 87.95 % and F1-Score of 91.47%.