Downloads: 2
Review Papers | Computer Science | Volume 15 Issue 2, February 2026 | Pages: 1709 - 1717 | India
Deep Learning for Myocardial Infarction Detection Using Echocardiography: A Comprehensive Survey of Segmentation and Classification Approaches
Abstract: Myocardial infarction (MI) arises when blood circulation to a segment of cardiac muscle is suddenly obstructed, resulting in tissue ischemia and irreversible myocardial damage. Imaging techniques, including Electrocardiography (ECG), Echocardiography, Coronary Angiography, and Computed Tomography (CT) play a vital role in assessing cardiac structure and function for MI diagnosis. Among these, echocardiography is widely used due to its non-invasive nature and ability to evaluate Regional Wall Motion Abnormalities (RWMA) and ventricular performance. However, manual evaluation of these images is labor-intensive and susceptible to inter-observer fluctuation, especially in detecting subtle wall motion changes. To overcome these challenges, Machine Learning (ML) techniques were initially introduced to support MI detection through automated segmentation and classification. But ML approaches relied heavily on manual feature engineering and often struggled with high-dimensional and complex patterns in echocardiographic sequences. In the last few years, Deep Learning (DL) techniques have surfaced as a formidable alternative which can autonomously learning classification characteristics using raw data. This survey reviews existing DL-based approaches for MI detection using echocardiography, thereby analyzing their methodologies, datasets, and performance metrics, and provides a comparative evaluation to identify the most reliable and clinically applicable frameworks.
Keywords: Myocardial infarction, Deep Learning, Echocardiography, Classification, Segmentation
How to Cite?: A. Shobana, K. P. Malarkodi, "Deep Learning for Myocardial Infarction Detection Using Echocardiography: A Comprehensive Survey of Segmentation and Classification Approaches", Volume 15 Issue 2, February 2026, International Journal of Science and Research (IJSR), Pages: 1709-1717, https://www.ijsr.net/getabstract.php?paperid=SR26227110915, DOI: https://dx.dx.doi.org/10.21275/SR26227110915