PUBLICATIONS
Dikici E, O'Donnell TP, Setser RM, White RD. Quantification of delayed enhancement MR images. Proceedings of Medial Image Computing and Computer Aided Intervention, 2004. St. Malo, France. (Winner Best Student Paper Image Segmentation and Processing Category) Vol 1: p. 250-257.
ABSTRACT:
Delayed Enhancement MR is an imaging technique by which non-viable (dead) myocardial tissues appear with increased signal intensity. The extent of non-viable tissue in the left ventricle ( LV ) of the heart is a direct indicator of patient survival rate. In this paper we propose a two-stage method for quantifying the extent of non-viable tissue. First, we segment the myocardium in the DEMR images. Then, we classify the myocardial pixels as corresponding to viable or non-viable tissue. Segmentation of the myocardium is challenging because we cannot reliably predict its intensity characteristics. Worse, it may be impossible to distinguish the infracted tissues from the ventricular blood pool. Therefore, we make use of MR Cine images acquired in the same session (in which the myocardium has a more predictable appearance) in order to create a prior model of the myocardial borders. Using image features in the DEMR images and this prior we are able to segment the myocardium consistently. In the second stage of processing, we employ a Support Vector Machine to distinguish viable from non-viable pixels based on training from an expert.
O'Donnell TP, Dikici E, Setser RM. Tracking and Analysis of Cine-Delayed Enhancement MR . Proceedings of Medial Image Computing and Computer Aided Intervention, 2005. Palm Springs, California, USA.
ABSTRACT:
Cine-DEMR is a new cardiac imaging technique which combines aspects of Cine and Delayed Enhancement MR. Like Cine, it displays the heart beating over time allowing for the detection of motion abnormalities. Like DEMR, non-viable (dead) tissues appear with increased signal intensity (it has been shown that the extent of non-viable tissue in the left ventricle ( LV ) of the heart is a direct indicator of patient survival rate). We present a technique for tracking the myocardial borders in this modality and classifying myocardial pixels as viable or non-viable. Tracking is performed using an affine deformed template of borders manually drawn on the first phase of the series and refined using an ASM-like approach. Classification employs a Support Vector Machine trained on DEMR data. We applied our technique on 75 images culled from 5 patient data sets.