Automated detection of airway and vessel orientations for quantitative analysis and visualization

1. A method, comprising:

displaying a three-dimensional (3D) image of a lung, wherein the 3D image includes a segmented bronchial tree;

receiving a selection of an airway of the lung, wherein the selected airway is identified by a click-point outside the segmented bronchial tree;

locally segmenting the selected airway, determining its orientation/direction and computing a two-dimensional (2D) cross-section of the airway perpendicular to the airways's long axis; and

displaying a (2D) cross-section image of the airway perpendicular to the airway's long axis, wherein the display of the 2D cross-section image occurs immediately after the selection of the airway is received,

wherein the selected airway is locally segmented by:

creating a volume of interest (VOI) around a point of the 3D image, wherein the point identifies the selected airway;

performing a 3D segmentation of the airway within the VOI, wherein the 3D segmentation of the airway includes minimum and maximum thresholds based on a neighborhood of the click-point, wherein the minimum threshold is set to prevent the 3D segmentation from expanding into a bifurcation and the maximum threshold is a Hounsfield value corresponding to air; and

locally thinning the 3D segmentation of the airway to ensure that only one branch is segmented.

2. The method of claim 1 , wherein the long axis of the airway is determined by computing eigenvectors of a Hessian matrix of the segmented VOI around the point.

3. The method of claim 2 , wherein the 2D cross-section of the airway perpendicular to the airway's long axis is computed by using interpolation.

4. The method of claim 1 , wherein the long axis of the airway is determined by computing eigenvectors of a covariance matrix of the segmented VOI around the point.

5. The method of claim 4 , wherein the 2D cross-section of the airway perpendicular to the airway's long axis is computed by using interpolation.

6. The method of claim 1 , further comprising qualitatively evaluating the airway in the 2D cross-section image.

7. The method of claim 1 , further comprising quantitatively evaluating the airway in the 2D cross-section image.

8. The method of claim 1 , wherein the 3D image of the lung is a Computed Tomography (CT) image.

9. A method, comprising:

displaying a three-dimensional (3D) image of a lung, wherein the 3D image includes a segmented vessel tree;

receiving a selection of a vessel of the lung, wherein the selected vessel is identified by a click-point outside the segmented vessel tree;

locally segmenting the selected vessel, determining its orientation/direction and computing a two-dimensional (2D) cross-section of the vessel parallel to the vessel'long axis; and

displaying a (2D) cross-section image of the vessel parallel to the vessel's long axis, wherein the display of the 2D cross-section image occurs immediately after the selection of the vessel is received.

wherein the selected vessel is locally segmented by:

creating a volume of interest (VOI) around a point of the 3D image, wherein the point identifies the selected vessel;

performing a 3D segmentation of the vessel within the VOI, wherein the 3D segmentation of the vessel includes minimum and maximum thresholds based on a neighborhood of the click-point, wherein the minimum threshold is set to prevent the 3D segmentation from expanding into a bifurcation and maximum threshold is a Hounsfield value corresponding to blood; and

locally thinning the 3D segmentation of the vessel to ensure that only one branch is segmented.

10. The method of claim 9 , wherein the vessel includes an artery or a vein.

11. A computer program product, comprising:

a non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising:

computer readable program code configured to perform the steps of:

visualizing a three-dimensional (3D) segmented bronchial tree;

receiving a user input, wherein the user input corresponds to a click-point that identifies an airway and the click-point is outside of the 3D segmented bronchial tree;

locally segmenting the selected airway and determining its orientation/direction;

computing a two-dimensional (2D) cross-section of the airway perpendicular to the airway's long axis; and

displaying an image of the 2D cross-section of the airway perpendicular to the airway's long axis,

wherein the selected airway is locally segmented by:

creating a volume of interest (VOI) around a point of the 3D segmented bronchial tree, wherein the point identifies the selected airway;

performing a 3D segmentation of the airway within the VOI, wherein the 3D segmentation of the airway includes minimum and maximum thresholds based on a neighborhood of the click-point, wherein the minimum threshold is set to prevent the 3D segmentation from expanding into a bifurcation and the maximum threshold is a Hounsfield value corresponding to air; and

locally thinning the 3D segmentation of the airway to ensure that only one branch is segmented.

12. The computer program product of claim 11 , wherein the display of the 2D cross-section image occurs immediately after the selection of the airway is received.

13. The computer program product of claim 11 , wherein the computer readable program code is further configured to perform the step of measuring the airway.

14. The computer program product of claim 13 , wherein the measurements include the airway's inner diameter or outer diameter.