DocumentCode
3525305
Title
Fast extraction of planes normal to the centreline from CT colonography datasets
Author
Sezille, N. ; Sadleir, R.J.T. ; Whelan, P.F.
Author_Institution
Vision Syst. Lab., Dublin City Univ., Ireland
fYear
2003
fDate
7-9 July 2003
Firstpage
161
Lastpage
164
Abstract
CT colonography (CTC) is an emerging imaging technique for colon screening. It is possible to extend this technique to automatically detect colonic anomalies. In order to improve the detection rate of the potential polyps, it is desirable to analyse plane sections orthogonal to the colon centreline rather than the axial CT slices. This paper describes a fast artefact free method for orthogonal plane extraction. The centreline in its initial form lacks the smoothness required to provide accurate local orientation of the colon. An approach based on down-sampling and B-spline interpolation generates a suitable smooth centreline. The cubic B-spline form gives low cost access to accurate local orientation and curvature of the colon. The local orientation is used as a normal vector for the orthogonal plane. The plane extraction from the anisotropic CT dataset is divided into two stages. First, a projection of the orthogonal plane on one of the dataset faces is created. A bilinear interpolation on the projected plane extracts the actual viewing window. This plane extraction is fast: extracting a plane using trilinear interpolation requires ∼7*N2 linear interpolations. Our method achieves the same image quality and requires only ∼3*N2 linear interpolations. This technique can be use to analyse cross-sections of the colon without significant processing overhead.
Keywords
computerised tomography; edge detection; feature extraction; interpolation; medical image processing; splines (mathematics); B-spline interpolation; bilinear interpolation; centreline processing; colon; colonic anomalies; computed tomography colonography; oblique plane extraction; orthogonal plane extraction; polyp detection;
fLanguage
English
Publisher
iet
Conference_Titel
Visual Information Engineering, 2003. VIE 2003. International Conference on
ISSN
0537-9989
Print_ISBN
0-85296-757-8
Type
conf
DOI
10.1049/cp:20030512
Filename
1341317
Link To Document