Securing Cardiovascular Data: A Cybersecurity Framework for Myocardial Infarction Prediction Systems

Abstract

This paper proposes a cybersecurity framework to enhance security with no impact on functionality compared to prior Myocardial Infarction prediction systems. Tampering and integrity violations of the sensitive cardiovascular data will be protected from unauthorized access using AES-256 encryption and multi-factor authentication with anomaly detection. Myocardial Infarction prediction through the use of predictive models such as Random Forest, SVM, and logistic regression proves to be quite accurate and robust. Results showed that the framework can successfully avoid attacks under acceptable prediction accuracy while meeting both requirements of data security and system performance.