9 May 2002 Automated segmentation method for the 3D ultrasound carotid image based on geometrically deformable model with automatic merge function
Author Affiliations +
Abstract
Stenosis of the carotid is the most common cause of the stroke. The accurate measurement of the volume of the carotid and visualization of its shape are helpful in improving diagnosis and minimizing the variability of assessment of the carotid disease. Due to the complex anatomic structure of the carotid, it is mandatory to define the initial contours in every slice, which is very difficult and usually requires tedious manual operations. The purpose of this paper is to propose an automatic segmentation method, which automatically provides the contour of the carotid from the 3-D ultrasound image and requires minimum user interaction. In this paper, we developed the Geometrically Deformable Model (GDM) with automatic merge function. In our algorithm, only two initial contours in the topmost slice and four parameters are needed in advance. Simulated 3-D ultrasound image was used to test our algorithm. 3-D display of the carotid obtained by our algorithm showed almost identical shape with true 3-D carotid image. In addition, experimental results also demonstrated that error of the volume measurement of the carotid based on the three different initial contours is less that 1% and its speed was a very fast.
© (2002) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiang Li, Zigang Wang, Hongbing Lu, Zhengrong Liang, "Automated segmentation method for the 3D ultrasound carotid image based on geometrically deformable model with automatic merge function", Proc. SPIE 4684, Medical Imaging 2002: Image Processing, (9 May 2002); doi: 10.1117/12.467111; https://doi.org/10.1117/12.467111
PROCEEDINGS
6 PAGES


SHARE
Back to Top