1. Lung lobe and segmental lobe extraction from 3D chest CT datasets based on figure decomposition and Voronoi division
- Author
-
Yasuhito Suenaga, Yuichi Nakada, Kensaku Mori, Hirotsugu Takabatake, Takayuki Kitasaka, Hiroshi Natori, and Masaki Mori
- Subjects
Thorax ,medicine.medical_specialty ,Lung ,business.industry ,Chest ct ,Anatomy ,respiratory system ,Division (mathematics) ,Thresholding ,Lobe ,Lung lobe ,respiratory tract diseases ,medicine.anatomical_structure ,medicine ,Radiology ,Voronoi diagram ,business - Abstract
In this paper, we present a method for segmenting lung lobe regions and segmental lobe regions from 3D CT datasets. In a CAD system for the chest, it is very important to understand structures of the chest by a computer. It is required to develop algorithms that automatically segment organ regions of the chest area from input 3-D chest CT datasets. This paper tries to divide lung regions into lung lobe regions and segmental lobe regions by the figure decomposition process and the Voronoi division process. In this method, we enhance sheet structures on CT images by using eigenvalues of a Hessian matrix. Also, the lung regions are segmented by simple thresholding and morphological filtering. Then, we subtract sheet structures from the lung regions. By applying the figure decomposition process, we obtain each lung lobe region. Segmental lobe regions are obtained by the Voronoi division process using bronchial branch information The proposed method was applied to thirteen cases of 3-D chest CT images. Experimental results showed that the proposed method can extract lung lobe regions and segmental lobe regions even for cases of incomplete fissure or over-extraction of interlobar pleura.
- Published
- 2008