DocumentCode :
541682
Title :
Three-dimensional analysis of regional left ventricular endocardial curvature from cardiac magnetic resonance images
Author :
Maffessanti, F. ; Caiani, E.G. ; Nesser, H.J. ; Niel, J. ; Steringer-Mascherbauer, R. ; Lang, R.M. ; Mor-Avi, V.
Author_Institution :
Univ. of Chicago, Chicago, IL, USA
fYear :
2010
fDate :
26-29 Sept. 2010
Firstpage :
883
Lastpage :
886
Abstract :
Left ventricular (LV) remodeling is usually assessed using changes in LV volume, while disregarding regional changes that may occur independently of volume. We hypothesized that 3D analysis of regional endocardial curvature could provide useful information on localized remodeling. Cardiac magnetic resonance (CMR) images were acquired in 44 patients: 14 normal controls (NL), 15 with dilated cardiomyopathy (DCM), and 15 ischemic heart disease (IHD). LV endocardial surface was reconstructed throughout the cardiac cycle and used to calculate for each point the curvedness, normalized by instantaneous LV size (Cn). Normalized curvedness was compared between groups of segments: NL (N=401), DCM (N=255) and IHD (N=92). While in NL segments, Cn values were comparable in basal and mid-ventricular segments, they were higher in the apical segments (p<;0.05). Also, % change in Cn was higher in mid and apical compared to basal segments (p<;0.05). At all LV levels, Cn in DCM segments was lower (p<;0.05) than in NL and IHD. In contrast, % change in Cn was lower in both IHD and DCM segments compared to NL (p<;0.05). 3D analysis of regional LV endocardial curvature from CMR images provides quantitative information, which is consistent with the known pathophysiology, and may prove useful in the evaluation of LV remodeling.
Keywords :
biomedical MRI; cardiovascular system; diseases; image classification; image segmentation; medical image processing; muscle; physiological models; 3D curvature analysis; apical segments; basal segments; cardiac cycle; cardiac magnetic resonance images; dilated cardiomyopathy; endocardial surface; ischemic heart disease; pathophysiology; regional left ventricular endocardial curvature; segment classification; three-dimensional analysis; ventricular segments; Cardiology; Image reconstruction; Image segmentation; Motion segmentation; Shape; Surface reconstruction; Three dimensional displays;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computing in Cardiology, 2010
Conference_Location :
Belfast
ISSN :
0276-6547
Print_ISBN :
978-1-4244-7318-2
Type :
conf
Filename :
5738115
Link To Document :
بازگشت