Table of Contents Author Guidelines Submit a Manuscript
International Journal of Biomedical Imaging
Volume 2012 (2012), Article ID 382806, 9 pages
Research Article

Automated Lobe-Based Airway Labeling

1Department of Radiology, University of Pittsburgh, Pittsburgh, PA 15213, USA
2Department of of Pharmacology & Chemical Biology, University of Pittsburgh, Pittsburgh, PA 15213, USA
3Department of Medicine, University of Pittsburgh, Pittsburgh, PA 15213, USA
4Department of Epidemiology, University of Pittsburgh, Pittsburgh, PA 15213, USA
5Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA 15213, USA

Received 6 April 2012; Revised 6 September 2012; Accepted 9 September 2012

Academic Editor: Ayman El-Baz

Copyright © 2012 Suicheng Gu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Regional quantitative analysis of airway morphological abnormalities is of great interest in lung disease investigation. Considering that pulmonary lobes are relatively independent functional unit, we develop and test a novel and efficient computerized scheme in this study to automatically and robustly classify the airways into different categories in terms of pulmonary lobe. Given an airway tree, which could be obtained using any available airway segmentation scheme, the developed approach consists of four basic steps: (1) airway skeletonization or centerline extraction, (2) individual airway branch identification, (3) initial rule-based airway classification/labeling, and (4) self-correction of labeling errors. In order to assess the performance of this approach, we applied it to a dataset consisting of 300 chest CT examinations in a batch manner and asked an image analyst to subjectively examine the labeled results. Our preliminary experiment showed that the labeling accuracy for the right upper lobe, the right middle lobe, the right lower lobe, the left upper lobe, and the left lower lobe is 100%, 99.3%, 99.3%, 100%, and 100%, respectively. Among these, only two cases are incorrectly labeled due to the failures in airway detection. It takes around 2 minutes to label an airway tree using this algorithm.