HRIDAYA- A Cardiovascular Disease Prediction Model

Non-communicable diseases pose a significant threat to human life especially in middle-income countries; one such class of disease is Cardiovascular Diseases (CVDs) which have been the leading cause of death worldwide. As per the numbers reported, there has been a global jump of 60% in the number of deaths caused by CVDs. Thus, this research aims to develop a solution using advanced technologies like machine learning and data science, that can accurately predict a possible CVD in a patient based on their clinical parameters so that precautions can be taken to evade health risks. There has been plenty of research in this domain of study where various data mining and machine-learning techniques have been used. The techniques include the usage of classifiers like Logistic Regression, Support Vector Machine, and Decision Trees. There has also been some research revolving around the linear correlation between clinical parameters and CVD, feature selection techniques, bagging, and boosting. However, to address this problem more efficiently we propose Hridaya - a cardiovascular disease detection model that has been developed using advanced techniques like ensemble learning and k-fold cross-validation which can easily identify hidden patterns in the given data. The focus has been to improve the model’s accuracy as well as recall because increased False Negatives can be fatal in a disease prediction model. The model was able to give a recall of 94% and an accuracy of 91% demonstrating the robustness of the model.