A “loop” shape descriptor and its application to automated segmentation of airways from CT scans

摘要: Purpose: A novel shape descriptor is presented to aid an automated identification of the airways depicted on computed tomography (CT) images. Methods: Instead simplifying tubular characteristic as ideal mathematical cylindrical or circular shape, proposed “loop” exploits fact that cross sections any structure (regardless its regularity) always appear a loop. In implementation, authors first reconstruct anatomical structures in volumetric CT three-dimensional surface model using classical marching cubes algorithm. Then, loop applied locate with concave section. To deal variation airway walls density images, multiple threshold strategy proposed. publicly available chest database consisting 20 scans, which was designed specifically for evaluating segmentation algorithm, used quantitative performance assessment. Measures, including length, branch count, and generations, were under skeletonization operation. Results: For test dataset, length ranged from 64.6 429.8 cm, generation 7 11, number 48 312. These results comparable state-of-the-art algorithms validated same dataset. Conclusions: The authors’ experiment demonstrated feasibility reliability developed identifying lung airways.