Abstract

High resolution CT datasets from multi-slice scanners allow the inspection of thetracheobronchial tree down to the smaller airways for diagnostic purposes (e.g. to inspectairway lumen and bronchial wall thickness). Using manual or automated segmentationmethods, the bronchial tree structure may be discernable down to hundreds of subsegmentsof higher generations. Linear projections (such as maximum intensity projections) of thedelicate three-dimensional tree lead to multiple occlusions (left figure), and make theinspection and navigation along specific airway paths cumbersome.The planar mapping is computed automatically for each automatically extractedindividual bronchial tree. It gives an overview of the bronchial tree in a single glance, isprintable (e.g. for the diagnostic report), allows to convey diagnostic CAD-information tothe reader through color-coding (e.g. anomalies in wall thickness or broncho-arterialdiameter ratios), and eases three-dimensional navigation through the bronchial tree (themap is bijectively linked to the 3D image volume).The suggested mapping is also applicable to show anomalies in the pulmonary arterialtree.Two possible approaches for automatic production of a non-intersecting topological mapping:multi-slice CT lung segmentation airway tree segmentation centerline extractionThe topological mapping (subway map) can aid the navigationthrough the bronchial tree. A mouse-click into the subway map willset the ortho-viewer to the corresponding true 3D-position, indicatethe position in the three-dimensional tree model, and show areformat of the selected airway location which is reformattedperpendicular to the local airway orientation .(a) At each bifurcation of generation i, the mappingbranches by an angle ± θ

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call