- Poster presentation
- Open Access
Fully automatic segmentation of short and long axis cine cardiac MR
© Fradkin et al; licensee BioMed Central Ltd. 2009
- Published: 28 January 2009
- Left Ventricle
- Automatic Segmentation
- Segmented Contour
- Left Ventricle Volume
- Left Ventricle Myocardium
Quantitative analysis of cardiac function requires delineation of the left ventricle (LV) in cine cardiac MR (CMR). Typically, this is done using short-axis (SA) images, however, acquisition of several long-axis (LA) views has become quite common. The latter can be used for the accurate and reproducible determination of the basal SA slice, known as one of the major inter-observer variability factors in SA LV measurements . Since manual LV delineation is very tedious and time-consuming, automatic segmentation methods, enabling to obtain reproducible LV measurements, are highly desirable.
We propose a fully automatic method for delineation of the endo- and epicardial contours in SA and LA cine CMR images in order to provide automatic, accurate quantitative left-ventricular functional assessment.
The initial position of the LA template is derived from the image acquisition geometry, while that for the SA templates is obtained by projecting the LA segmented contours on the corresponding SA slices.
We quantitatively assessed the performance of the method on a database of 35 cine CMR acquisitions (Philips Gyroscan NT Intera 1.5 T, SSFP protocol, TE = 1.6 ms, TR = 3.3 ms, flip angle 50° for SA and TE = 1.5–1.7 ms, TR = 3.0–3.4 ms, flip angle 55° for LA). The SA ones included 9 to 14 slices (325 slices in total), and the LA ones 1 to 2 views (40 views in total). The assessment was made by comparing the automatically segmented contours with Gold Standard manual delineations provided by an expert.
We presented a robust, accurate and efficient method for the fully automatic delineation of the myocardium contours in LA + SA cine CMR images, which can be used for accurate LV functional assessment.
- Weaver A, et al: J Cardiovasc Magn Reson. 2006, 1: 82-83.Google Scholar
- Williams DJ, Shah M: CVGIP: Image Understanding. 1992, 55 (1): 14-26. 10.1016/1049-9660(92)90003-L.View ArticleGoogle Scholar
- Hautvast , et al: IEEE Trans Med Imag. 2006, 25 (11): 1472-1482. 10.1109/TMI.2006.882124.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd.