A novel device to detect myocardial infarction

Myocardial infarction, commonly known as heart attack, is one of the major causes of death around the world. For many, heart attacks are unexpected and can occur at any time, especially if a person previously had a heart attack or any type of heart disease. The suddenness of a heart attack makes it difficult to detect and prevent it from occurring, resulting in death or irreversible injury to the heart. Finding a method of detecting a heart attack even five minutes before the attack occurs can be the time between life and death. My research aims to use a machine learning algorithm incorporated into a noninvasive biosensor for early detection of heart attacks. Users first enter factors such as biometrics, history of cardiac diseases, and habits. The biosensor will have a live feed of ECG data from the user. The neural network algorithm will take these initial factors as well as the ECG data to determine whether or not a user is experiencing a myocardial infarction. The neural network is trained by data from the PTB Diagnostic ECG Database from PhysioNet. This project will allow early detection of a heart attack, thus early treatment and a decreased possibility of death and long term tissue damage, and can also be used to track user heart health over a period of time.