|
KMID : 1147120070130010019
|
|
Journal of the Korean Society of Imaging Informatics in Medicine
2007 Volume.13 No. 1 p.19 ~ p.25
|
|
|
Automatic Airway Segmentation in MDCT using Intensity Adaptive Region-growing Technique
|
|
Park Sang-Jun
Heo Chang-Yong
Lee Hyun-Ju
Goo Jin-Mo
Kim Jong-Hyo
|
|
|
Abstract
|
|
|
|
For efficient management of asthma and Chronic Obstructive Pulmonary Diseases (COPD), quantitative assessment of airway geometry is necessary. However caused by technical difficulties such as partial volume effect of CT (computed tomography) and limitation in image quality inherent to CT image acquisition make airway analysis difficult. Besides, airway tree segmentation in CT images is a challenging problem because of the complex anatomy. Furthermore, the limited intensity contrast between the participating materials (air, blood, and tissue) increases the segmentation difficulties. In this reason, we have developed a three-dimensional segmentation technique of airways with a proposed algorithm. Firstly, a seed point is automatically identified by searching the air-filled region near the center of the first few slices in the data set. Then, we can segment accurate bronchial tree using our proposed method with maximizing detection of airway intensity range. Using CT scans of human cases, we can understand and appreciate our proposed method is convenient and accurate for quantitative measurement.
|
|
|
KEYWORD
|
|
|
Computed tomography (CT), Three-dimensional, Airway, Segmentation, Image processing
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|