Classifications

• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/10—Segmentation; Edge detection
  • G06T7/11—Region-based segmentation
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10072—Tomographic images
  • G06T2207/10081—Computed x-ray tomography [CT]
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/30—Subject of image; Context of image processing
  • G06T2207/30004—Biomedical image processing
  • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Definitions

• This invention pertains in general to the field of medical imaging. More particularly the invention relates to a method and an apparatus for planar angular visualization of an anatomical tree structure, and more particularly to a planar topological mapping of a three-dimensional bronchial tree structure of the lungs.
• CT computed tomography
• a patient is advanced linearly through the examination region along a direction that is perpendicular to a gantry rotation plane to effectuate a helical orbiting of the x-ray source about the subject.
• X-ray absorption data obtained during the helical orbiting is reconstructed using filtered back projection or another reconstruction method to generate a three-dimensional image representation of the subj ect or of a selected portion thereof.
• High resolution CT data sets from multi-slice scanners allow the inspection of the bronchial down to smaller airways for diagnostic purposes.
• the resulting images can be used to inspect the airway lumen and wall thickness of the bronchia.
• the bronchial tree structure contains hundreds of sub-segments. Therefore, a visual inspection of all paths in the three-dimensional bronchial tree is very time consuming. On the other hand, a planar projection of the three-dimensional tree leads to multiple occlusions, as illustrated in Figure 1, and the continuation of the specific airway path is difficult to follow, due to the delicate structure of the tree.
• the present invention preferably seeks to mitigate, alleviate or eliminate one or more of the above-identified deficiencies in the art and disadvantages singly or in any combination and solves at least the above mentioned problems by providing a system, an image acquisition device, a method, a computer-readable medium and an image workstation that provides a planar angular visualization of an anatomical tree according to the appended claims.
• the general solution according to the invention suggests a planar topological mapping of the three-dimensional bronchial tree structure of the lungs such that no occlusions or intersection of segments occur, that the hierarchy of the sub-segments is derivable, and that conveys a similarity to the true anatomical regions such as the lung lobes.
• this map allows inspection of the bronchial tree in a view, and with a low degree of freedom, for instance through a Graphical User Interface slider which varies the selected distance to the carina of all bronchial segments.
• a method for visualizing an anatomical tree structure which comprises the following steps.
• the anatomical tree structure is segmented from a three-dimensional image set of at least a portion of a body.
• a planar angular visualization of the anatomical tree structure is then determined.
• the planar angular visualization of the anatomical tree structure is then displayed.
• a system for visualizing an anatomical tree structure is disclosed.
• Segmenting means segments the anatomical tree structure from a three-dimensional image set of at least a portion of a body.
• Determination means determines a planar visualization of the anatomical tree structure.
• Displaying means displays the planar angular visualization of the anatomical tree structure.
• a computer-readable medium having embodied thereon a computer program, for automatic extraction of an anatomical tree structure, for processing by a computer.
• the computer program comprises a code segment for segmenting the anatomical tree structure from a three-dimensional image set of at least a portion of a body; a code segment for determining a planar angular visualization of the anatomical tree structure; and a code segment for displaying the planar angular visualization of the anatomical tree structure.
• the present invention has the advantage over the prior art that it provides a two-dimensional representation of an anatomical map in which occlusions or intersection of segments are minimized, i.e. do not occur, thus making the visual inspection of the paths of the anatomical tree much easier, thus facilitating subsequent automated or manual processing of the derived information, e.g. for analysis of the tree for anomalies.
• Fig. 3 is a flow chart illustrating the method for creating a planar angular visualization of an anatomical tree structure according to one embodiment of the invention
• Fig. 4 illustrates a planar angular mapping of the bronchial tree extracted from a particular CT lung data set according to one embodiment of the invention
• Fig. 6 illustrates various reformatted images according to one embodiment of the invention
• Fig. 7 illustrates an alternative display of the planar mapping of the bronchial tree as extracted from a particular CT lung data set according to one embodiment of the invention
• Fig. 8 schematically illustrates a computer readable medium according to one embodiment of the invention.
• CT computed tomography
• MRI Magnetic Resonance Imaging
• PET Position Emission Tomography
• SPECT Single Photon Emission Computed Tomography
• the CT acquisition device 20 comprises a multi-array CT gantry 22 and a patient table 24 that can be positioned within the gantry 22.
• the patient table 24 supports a patient during acquisition of raw image data of the patient.
• the coarse image data is applied to a computer 26, which reconstructs volumetric image data of the raw image data.
• the computer 26 is programmed in such a manner that in conformity with the invention the computer 26 calculates a segmented tree of the tracheobronchial tree.
• the computer 26 determines a planar angular visualization of the anatomical tree structure, the tree and visualization being displayed on a display unit of the computer.
• the workstation 30 comprises a suitable storage reading device, like a CD-drive, that can read the software from the storage device. This CD-drive is then operatively connected to the software bus too.
• a person is x-rayed using a multi-slice CT and a 3-D data set representing the person's bronchial tree is created using a known manner in step 300.
• the bronchial tree is segmented from a three-dimensional image in step 302, as for instance based on the algorithm described in Schlat ⁇ lter et al. (Simultaneous Segmentation and Tree Reconstruction of the Airways for Virtual Bronchoscopy, SPIE Conference on Medical Imaging, Proceedings of SPIE Vol. 4684, pp. 103-113 (2002)).
• the unsupervised bronchial tree segmentation starts from the trachea and is based on a so-called front propagation approach.
• a region growing aggregates all voxels below a certain Hounsfield threshold. Whenever the growth front splits into different segments, a further region growing is started in each of the segments. "Leakages" into the parenchymal tissue is detected and stopped by monitoring a possible sudden increase of the growth front. After the region growing process is complete, then the centerlines (skeleton lines) of trachea, bronchi and smaller airways are extracted and a linked graph structure is built which represents the branching points of the tree structure.
• a planar angular visualization of the bronchial tree structure is then determined in step 304.
• the planar visualization of the bronchial tree structure is then displayed in step 306.
• the invention suggests a two-dimensional planar mapping of the bronchial tree, as illustrated in Figure 4, such that no occlusions or intersection of segments occur, and that the hierarchy of the segments is obvious.
• the true geometric relations are not preserved, but the topological relations are true as for example in a typical subway map.
• the main segments point in the approximate direction of their true geometric location in a coronal projection wherein upper lung lobes (42, 41) are directed upwards and lower lung lobes (44, 43) are directed downwards as illustrated in Figures 4-5.
• Figure 5 also illustrates a middle 45 lung lobe.
• the suggested planar mapping gives an overview of the bronchial tree in a single two-dimensional view, which is intuitively similar to the anatomical relations, as illustrated in Figure 5.
• the main bifurcation (carina) in the trachea 50 into left 52 and right 54 main bronchus is plotted in the center of the planar diagram as illustrated in Figure 4. All other points in the diagram are plotted such that their radial distance to the diagram center is proportional to the centerline distance through the airways to the carina, measured in millimetres independent of CT data set resolution.
• the trachea 50 descends towards the center of the diagram, representing the carina, and then left and right main bronchi bifurcate at 90° angles. It will be understood by those skilled in the art that the angle could also be fixed at another value, for example, as illustrated in Figure 7 which will be described in more detail below.
• the plotted centerline branches with an equal angular step width to the left and right of the current angular direction.
• the angular step width decreases by a factor of 1 A after each bifurcation.
• the bifurcations are characterized as the locations where in the above described region growing scheme the growth front has disintegrated (split). In this way, it can be guaranteed that the subtree below a given point will not enter the diagram area of another subtree.
• Each point in the mapping is given by angle and distance to the center of the diagram.
• the distance to the diagram center is proportional to the bronchial-path-distance to the carina, and the angle is determined by the number of previous bifurcations. So the length of a segment in the diagram is just given by the connection between the point positions determined as previously explained.
• the length of a segment in the diagram is not a true representation of its real length. This is not possible as the segments have to be spread out far enough to avoid overlapping. Rather, the radial distance is a true representation of the length of the airway back to the carina. In other words, the tangential distances in the diagram (on the circles) are just used to spread out the segments, but the radial distances are true.
• the direction of the branching in the diagram is controlled such that the one of the two subtrees which has the higher centroid (in foot-to-head direction) is chosen as the branch in the diagram which bends towards the top, whereas the other branch bends towards the bottom of the diagram.
• the colors or grey scale values of the displayed diagram may be chosen such as to reflect the anatomically differentiated lobes and main segments as illustrated in Fig. 10. It will be understood by those skilled in the art that the actual displayed colors or grey scale values can be selected by the user to provide a visual image which is distinctive to the user.
• the diagram may be interactive in a way that a mouse click on a given point in the diagram will set another display device such as an orthoviewer directly to the place in the three-dimensional CT data set which is represented by the click-on point in the planar mapping.
• the diagram can be drawn such that smooth angular interpolation is used to give a more organic appearance as illustrated in Fig. 7.
• the bifurcations are at the same positions as in Figs. 4-5, wherein position is defined by the number of earlier bifurcations and length to carina, but between the bifurcation points the centreline points are linearly interpolated in their angle and radial distance (p and ⁇ coordinate).
• the diagram can be drawn such that the bronchial wall thickness at each point of the diagram is proportional to the measured wall thickness at the corresponding point in the bronchial tree.
• Anomalies in the bronchial lumen diameter and wall thickness can be automatically detected and pointed out by markers or color coding at the respective positions in the diagram.
• another image processing module can find at each point of the bronchial tree the accompanying artery. The diameter of the accompanying artery can be automatically measured. Then the ratio between the bronchial diameter and the arterial diameter (which is an important clinical parameter) can be color-coded into each point of the planar mapping.
• planar mapping is to map the segmented pulmonary arterial vessel tree, and to render at each point of the graph the vessel diameter or any possibly detected anomalies. For example, at each point of the graph, the minimum projection of the CT-Hounsfield values at this vessel portion can be rendered, so that filling defects (caused by clots, and possibly indicating pulmonary embolisms) can be detected at a single glance.
• a computer readable medium is illustrated schematically.
• a computer-readable medium 100 has embodied thereon a computer program 110 for automatic extraction of a planar angular visualization of an anatomical tree from a 3D medical image 112, for processing by a computer 113.
• the computer program comprises a code segment 114 for segmenting the anatomical tree structure from a three-dimensional image set of at least a portion of a body, a code segment 115 for determining a planar angular visualization of the anatomical tree structure, a code segment 116 for displaying the planar angular visualization of the anatomical tree structure.
• Fig. 9 illustrates an exemplary medical image workstation 119 according to a further embodiment of the present invention.
• the medical workstation is arranged for implementing the method of the invention, and configured to receive and process a 3D medical image.
• the workstation is arranged to run the above described program code segments in order to perform the method according to the invention.

Landscapes

• Engineering & Computer Science (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Theoretical Computer Science (AREA)
• Apparatus For Radiation Diagnosis (AREA)

Priority Applications (4)

Applications Claiming Priority (4)

Publications (1)

Family

ID=35814189

Family Applications (1)

Country Status (5)

Cited By (2)

Families Citing this family (19)

Citations (3)

Family Cites Families (7)

• 2005
  • 2005-11-21 CN CN2005800401985A patent/CN101065771B/zh not_active Expired - Lifetime
  • 2005-11-21 JP JP2007542441A patent/JP4871292B2/ja not_active Expired - Fee Related
  • 2005-11-21 EP EP05806939.4A patent/EP1817742B1/en not_active Expired - Lifetime
  • 2005-11-21 WO PCT/IB2005/053829 patent/WO2006056923A1/en not_active Ceased
  • 2005-11-21 US US11/719,549 patent/US8045769B2/en active Active

Patent Citations (3)

Non-Patent Citations (2)

Also Published As

Similar Documents

Legal Events