Two Statistical Mixture Model vs. Fuzzy C-Means: In the application of edema segmentation

Abstract

Evaluating salvageable myocardial after myocardial infarction (MI) is an important prognosis in the follow up study of MI. Since the extent of myocardial edema delineates the ischemic area-at-risk (AAR) after MI the AAR can be used to estimate the amount of salvageable myocardial post-MI and therefore has potential clinical utility in the management of acute MI patients. Two methods for the segmentation and quantification of edema from T2 weighted MRI data have been presented. The methods presented in this paper are Two Statistical Mixture Model and Fuzzy C-means. Quantitative evaluations of segmentation accuracy for the two algorithms were performed by comparing to manual segmentation on real T2 weighted CMR data collected from Golden Jubilee National Hospital, Glasgow for 16 adult subjects.