Fully automated 2D and 3D convolutional neural networks pipeline for video segmentation and myocardial infarction detection in echocardiography

Cardiac imaging known as echocardiography is a non-invasive tool utilized to produce data including images and videos, which cardiologists use diagnose cardiac abnormalities in general myocardial infarction (MI) particular. Echocardiography machines can deliver abundant amounts of that need be quickly analyzed by help them make diagnosis treat conditions. However, the acquired quality varies depending on acquisition conditions patient's responsiveness setup instructions. These constraints are challenging doctors especially when patients facing MI their lives at stake. In this paper, we propose an innovative real-time end-to-end fully automated model based convolutional neural networks (CNN) detect regional wall motion (RWMA) left ventricle (LV) from videos produced echocardiography. Our implemented pipeline consisting 2D CNN performs preprocessing segmenting LV chamber apical four-chamber (A4C) view, followed 3D binary classification if segmented shows signs MI. We trained both CNNs dataset composed 165 each distinct patient. The achieved accuracy 97.18% segmentation while 90.9% accuracy, 100% precision 95% recall detection. results demonstrate creating system for detection feasible propitious.