Why the study?
Can an automatic detection algorithm using curvature-based features and a support vector machine reliably detect left ventricular aneurysms in echocardiograms?
Can an automatic detection algorithm using curvature-based features and a support vector machine reliably detect left ventricular aneurysms in echocardiograms?
An automated algorithm using curvature-based features and a support vector machine can reliably detect left ventricular aneurysms from 4-chamber echocardiogram views.
May facilitate automated LV aneurysm screening on echocardiography; provides first RCT validation of curvature-based ML for structural detection.
Left ventricular cardiac aneurysms are bulgings in the myocardium muscle of the left ventricle. These irregular distortions of the left ventricular shape from its normal bullet-like appearance, are often due to result of myocardial infarction and can be fatal. In this paper we address, for the first time, the automatic detection of left ventricular (LV) cardiac aneurysms from 4-chamber views in echocardiograms. For this, we first detect the left ventricle in the echocardiogram image as the lumen region closest to the apex of the heart. The apex itself is estimated from the bounding lines of the viewing sector in an echocardiogram. The boundary of the LV is then analyzed to extract key curvature-based features for discrimination using a support vector machine with radial basis function kernels. Results of testing on a large echocardiogram video collection indicate that robust detection of left ventricle coupled with curvature features is sufficient to reliably separate LV aneurysms from normal left ventricular shapes.
No takes yet. Share an insight, caveat, or question.
Mahmood et al. (2015) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: