Cardiovascular Prediction System Using Machine Learning
Pooja Gole1, Akshada Dhumal2, Piyush Jadhav3
Department of Computer Engineering, Narhe
Abstract
A heart attack is another term for a cardiovascular stroke. According to the World Health Organization, stroke is the second leading cause of death worldwide, accounting for 11% of all deaths. As a result of the high number of COVID-19 patients who experience breathing issues as a side effect of therapy, the risk of a stroke has increased. There isn’t always a system in place to check for the odds of having a stroke. A modest fluctuation in blood pressure that we underestimate can sometimes lead to a stroke. So, we created a system that will predict the possibility of having a stroke. We experimented with various of machine learning techniques to come up with the optimal of the solution. We chose the best performing algorithm in terms of accuracy and type-1 errors among the majority of machine learning algorithms that we tested with implementation. We devised a method for predicting the likelihood of suffering a stroke based on a few simple characteristics that maybe measured at home. As a result, if our algorithm predicts that you may have a stroke, you should seek medical advice as soon as possible. The chosen model is trained using a training set and evaluated using appropriate metrics such as accuracy, precision, recall, and AUC-ROC. The final model is deployed for predicting heart attacks on new, unseen data. The results demonstrate promising performance, with high accuracy and reliable prediction capabilities. The developed model holds significant potential for assisting healthcare professionals in early detection and prevention of heart disease, thereby improving patient care and reducing mortality rates. Future work may focus on expanding the dataset, incorporating additional features, and exploring advanced machine learning techniques to further enhance the predictive capabilities of the model.
Keywords: AWS, Cardiovascular Disease Prediction, Machine Learning Techniques, Random Forest model.