Share Email Print

Proceedings Paper

Adaptive model based pulmonary artery segmentation in 3D chest CT
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The extraction and analysis of the pulmonary artery in computed tomography (CT) of the chest can be an important, but time-consuming step for the diagnosis and treatment of lung disease, in particular in non-contrast data, where the pulmonary artery has low contrast and frequently merges with adjacent tissue of similar intensity. We here present a new method for the automatic segmentation of the pulmonary artery based on an adaptive model, Hough and Euclidean distance transforms, and spline fitting, which works equally well on non-contrast and contrast enhanced data. An evaluation on 40 patient data sets and a comparison to manual segmentations in terms of Jaccard index, sensitivity, specificity, and minimum mean distance shows its overall robustness.

Paper Details

Date Published: 12 March 2010
PDF: 9 pages
Proc. SPIE 7623, Medical Imaging 2010: Image Processing, 76234S (12 March 2010); doi: 10.1117/12.843750
Show Author Affiliations
Marco Feuerstein, Nagoya Univ. (Japan)
Takayuki Kitasaka, Aichi Institute of Technology (Japan)
Nagoya Univ. (Japan)
Kensaku Mori, Nagoya Univ. (Japan)

Published in SPIE Proceedings Vol. 7623:
Medical Imaging 2010: Image Processing
Benoit M. Dawant; David R. Haynor, Editor(s)

© SPIE. Terms of Use
Back to Top