Pulmonary Vasculature Segmentation¶

CIRRUS Lung includes a fully automatic segmentation of the pulmonary vasculature to allow for analysis of vacular abnormalities. The vasculature segmentation uses a Hessian matrix analysis based vesselness filter and includes a local thresholding procedure to accurately segment vessels of varying diameters. Since vesselness filters are known to often falsely include airway walls, the airway segmentation is used to remove these airway walls from the vessel segmentation. The segmentation of the pulmonary vasculature is performed offline and can be inspected in CIRRUS Lung as a 3D overlay depicting the vasculature on the original CT scan.

Technical publications about pulmonary vasculature segmentation algorithms included in CIRRUS Lung¶

E. van Dongen and B. van Ginneken. "Automatic segmentation of pulmonary vasculature in thoracic CT scans with local thresholding and airway wall removal", in: IEEE International Symposium on Biomedical Imaging, 2010, pages 668-671. Abstract/PDF DOI