A study of AAA image segmentation technique using geometric active contour model with morphological gradient edge function.

H. C. Kim, Y. H. Seol, S. Y. Choi, J. S. Oh, M. G. Kim, K. Sun

Research output: Contribution to journalArticlepeer-review

Abstract

Abdominal aortic aneurysm (AAA) is a serious vascular disease that can be life threatening. Accurate measurement of AAA size is important for surgical or endovascular repair. We have examined the feasibility of using the proposed method to drive quantitative measurement of a region of interest from AAA. The proposed geometric active contour model (PGACM) is a modification of the conventional geometric active contour model (CGACM) that uses morphological gradient edge function rather than Gaussian filtered images. The rationale for this is to eliminate the blurring effect induced by the Gaussian filter in the CGACM. We used three noised synthetic images with different shapes. To test performance, three quantities that were normalized for minimum distance error, mismatched area, and execution time are evaluated. PGACM, parametric active contour model (PACM), and CGACM were compared with respect to the three quantities. With PGACM, we obtained better performance for the segmentation than with the PACM and CGACM. This study shows the feasibility, accuracy, and precision of segmentation of AAA from CT data, and indicates that the proposed method may be useful in patients with AAA.

Original languageEnglish
Pages (from-to)4437-4440
Number of pages4
JournalConference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
Publication statusPublished - 2007 Dec 1
Externally publishedYes

ASJC Scopus subject areas

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Fingerprint

Dive into the research topics of 'A study of AAA image segmentation technique using geometric active contour model with morphological gradient edge function.'. Together they form a unique fingerprint.

Cite this