Towards myocardial infarction diagnosis: an investigation into machine learning and deep learning algorithms using ECG data

Abstract

Myocardial Infarction (MI) is a severe and potentially fatal heart condition caused by heart muscle injury due to lack of blood supply. Early detection and medical intervention can significantly reduce the mortality rate of MI. Electrocardiography (ECG) signals are commonly used to diagnose MI, but the process is susceptible to human error. This research aims to utilize machine learning and deep learning techniques to detect MI based on ECG signals and other biophysical factors. The incorporation of these factors tend to improve the predicted accuracy and resilience of the model. The performance of these machine learning and deep learning methods will depend on the carefully selected features and ECG signals. The anticipated result is expected to be a robust ML and DL-based diagnostic algorithm that can adapt to individual patient profiles, offering distinctive personalized risk assessments and magnifying the early detection of possible MI cases. Using machine learning to detect MI can enhance the accuracy and efficiency of diagnosis and identify individuals at high risk of developing MI for preventative measures. Moreover, using Deep Learning techniques that utilize image processing and visualization can help us understand why the model is making a decision. This decision taking pattern can later help us narrow down the type of MI the patient has. This study, overall, emphasizes the potential of machine learning and deep learning to improve our understanding of the relationship between medical conditions, biophysical factors and disease development.