Artificial intelligence-based myocardial infarction diagnosis: a comprehensive review of modern techniques

Myocardial infarction (MI), commonly known as a heart attack, is a serious medical condition that can lead to congestive heart failure and even death. Prompt diagnosis and early intervention are essential for improving a patient's survival chances. Electrocardiography (ECG) is the most commonly used diagnostic method for MI, but other noninvasive imaging techniques and clinical parameters are also used. However, manual interpretation of these methods can result in potential inconsistencies among different observers. To address this issue, automated computer-aided diagnostic systems that utilize artificial intelligence (AI) have been developed. These systems use both machine learning (ML) and deep learning (DL) models to discriminate between MI and normal signals or subjects. In this review paper, we survey the current state-of-the-art methods in ML and DL-based MI detection approaches that are published from 2015 to the present. This review highlights the advantages and limitations of different AI-based approaches and provides insights into future directions for research in this field. The ultimate goal of these efforts is to improve the accuracy and efficiency of MI diagnosis and contribute to more efficient and timely diagnosis of MI patients.