胸部CT画像の計算機診断支援に関する研究 : ―気管支領域の認識―

Abstract

This paper describes a novel method for extracting bronchial regions from 3D chest CT images. Several groups have proposed methods for extracting bronchial regions by region growing algorithms. However, these methods don't work well in the case that tumors divide the bronchial region into several parts. We propose a method that can extract the bronchial region in such cases using local intensity structure analysis, CT value distribution and shape features. First, the proposed method extracts cylindrical structure based on analysis of Hessian matrix. Second, it extracts tube structure by a modified RRF to analyze CT value distribution and shape features. At last, it extracts the bronchial region by integrating regions extracted by Hessian matrix analysis and RRF. We applied the proposed method to four cases of 3D chest CT images and showed that the proposed method could extract bronchial regions in the case that a tumor divides the bronchial region.