21 March 2014 Development of automated extraction method of biliary tract from abdominal CT volumes based on local intensity structure analysis
Author Affiliations +
Abstract
In this paper, we propose an automated biliary tract extraction method from abdominal CT volumes. The biliary tract is the path by which bile is transported from liver to the duodenum. No extraction method have been reported for the automated extraction of the biliary tract from common contrast CT volumes. Our method consists of three steps including: (1) extraction of extrahepatic bile duct (EHBD) candidate regions, (2) extraction of intrahepatic bile duct (IHBD) candidate regions, and (3) combination of these candidate regions. The IHBD has linear structures and intensities of the IHBD are low in CT volumes. We use a dark linear structure enhancement (DLSE) filter based on a local intensity structure analysis method using the eigenvalues of the Hessian matrix for the IHBD candidate region extraction. The EHBD region is extracted using a thresholding process and a connected component analysis. In the combination process, we connect the IHBD candidate regions to each EHBD candidate region and select a bile duct region from the connected candidate regions. We applied the proposed method to 22 cases of CT volumes. An average Dice coefficient of extraction result was 66.7%.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Kusuto Koga, Yuichiro Hayashi, Tomoaki Hirose, Masahiro Oda, Takayuki Kitasaka, Tsuyoshi Igami, Masato Nagino, Kensaku Mori, "Development of automated extraction method of biliary tract from abdominal CT volumes based on local intensity structure analysis", Proc. SPIE 9034, Medical Imaging 2014: Image Processing, 903448 (21 March 2014); doi: 10.1117/12.2043347; https://doi.org/10.1117/12.2043347
PROCEEDINGS
10 PAGES


SHARE
Back to Top