Quantification of the Spatial Distributionof Line Segments with Applications to CADof Chest X-Ray CT Images

We introduce two features to quantify distributions of line figures in the three-dimensional (3D) space. One of these is the Concentration index and the other is a feature based on the extended Voronoi tessellation. The former quantifies the degree of concentration, and the latter the difference of density. We explain the two features with their applications to the benign/malignant discrimination of lung tumors. The theoretical analysis is also shown.

[1]  Jun-ichi Hasegawa,et al.  Automated extraction of cancer lesions with convergent fold patterns in double contrast x-ray images of the stomach , 1991, Systems and Computers in Japan.

[2]  I. Ragnemalm.  Rotation invariant skeletonization by thinning using anchor points , 1993 .

[3]  Junichi Hasegawa,et al.  A new filter for feature extraction of line pattern texture with application to cancer detection , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol. III. Conference C: Image, Speech and Signal Analysis,.

[4]  Hironobu Ohmatsu,et al.  Quantification of shrinkage of lung lobe from chest CT images using the 3D extended Voronoi division and its application to the benign/malignant discrimination of tumor shadows , 2002, Object recognition supported by user interaction for service robots.

[5]  T. Saito,et al.  A sequential thinning algorithm for three dimensional digital pictures using the Euclidean distance transformation , 1995 .

[6]  Yoshito Mekada,et al.  A concentration index for analyzing three‐dimensional line patterns and its application to X‐ray CT images , 1998 .

[7]  Chris Pudney,et al.  Distance-Ordered Homotopic Thinning: A Skeletonization Algorithm for 3D Digital Images , 1998, Comput. Vis. Image Underst..

[8]  Alexis Gourdon,et al.  Computing the Differential Characteristics of Isointensity Surfaces , 1995, Comput. Vis. Image Underst..

[9]  Arie E. Kaufman,et al.  Fully automatic extraction of the colon centerline and its impact on a virtual colonoscopy system , 2001, CARS.

[10]  Rafael Wiemker,et al.  Optimal thresholding for 3D segmentation of pulmonary nodules in high resolution CT , 2001, CARS.

[11]  Yasushi Hirano,et al.  A tracking algorithm for extracting ridge lines in three-dimensional gray images using curvature of four-dimensional hypersurface , 2001, Systems and Computers in Japan.

[12]  George W.C. Hung,et al.  Stereological methods, vol. 2: Theoretical foundations: By Ewald R. Weibel. Academic Press, New York, 1980. xiv + 340 pp., $62.50 , 1984 .

[13]  Olivier Monga,et al.  Using Partial Derivatives of 3D Images to Extract Typical Surface Features , 1995, Comput. Vis. Image Underst..

[14]  Hironobu Ohmatsu,et al.  Extraction of tumor regions keeping boundary shape information from chest X-ray CT images and benign/malignant discrimination , 2001, CARS.

[15]  Ronald M. Summers,et al.  Grey-Scale Skeletonization of Small Vessels in Magnetic Resonance Angiography , 2000, IEEE Trans. Medical Imaging.

[16]  Yoshito Mekada,et al.  Three dimensional concentration index-a local feature for analyzing three dimensional digital line patterns and its application to chest X-ray CT images , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[17]  Richard A. Robb,et al.  Center line algorithm for virtual endoscopy based on chamfer distance transform and Dijkstra's single-source shortest-path algorithm , 1999, Medical Imaging.

[18]  Max A. Viergever,et al.  Fast delineation and visualization of vessels in 3-D angiographic images , 2000, IEEE Transactions on Medical Imaging.