Multi-structure atlas and/or use thereof

• US 9,672,614 B2
• Filed: 09/27/2013
• Issued: 06/06/2017
• Est. Priority Date: 10/09/2012
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Dwivedi, Shekhar
• Primary Examiner(s): Liew, Alex
• Application Number:
  US14/433,765
• Publication Number:
  US 20150243026A1
• Time in Patent Office: 1,348 Days
• Field of Search: 382128-134
• CPC Class Codes: G06T 11/005 Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/20104 Interactive definition of region of interest [ROI]
• CPC Class Codes: G06T 2207/20128 Atlas-based segmentation
• CPC Class Codes: G06T 2207/30008 Bone
• CPC Class Codes: G06T 2207/30061 Lung
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation

Abstract

An image data processor (106) includes a structural image data processor (114) that employs a multi-structure atlas to segment a region of interest from structural image data that includes tissue of interest and that segments the tissue of interests from the region of interest. The image data processor further includes functional image data processor (116) that identifies the tissue of interest in functional image data based on the segmented tissue of interest. An image data processor includes a multi-structure atlas generator (104) that generates a multi-structure atlas. The multi-structure atlas physically maps structure to tissue of interest such that locate the structure in structural image data based on the multi-structure atlas localizes the tissue of interest to the region of interest.

Claims

1. An image data processor, comprising:

a structural image data processor that employs a multi-structure atlas to segment a region of interest from structural image data which includes tissue of interest and that segments the tissue of interests from the region of interest, wherein the structure is rib cage and the tissue of interest is lung lobe, and the multi-structure atlas is a rib cage/lung lobe atlas that includes a mapping that maps one or more ribs to a location of a boundary of a lung lobe, and the mapping includes a lung lobe region of interest, in a form of a least one of a point cloud on a surface of the lung or a 3D sheet across lung, through the rib case/lung lobe atlas; and

a functional image data processor that identifies the tissue of interest in functional image data based on the segmented tissue of interest.

2. The image data processor of claim 1, wherein the functional image data processor further registers the structural image data and the functional image data and identifies the tissue of interest in the functional image data based on the segmented tissue of interest and the registered image data.

3. The image data processor of claim 1, wherein the functional image data processor further quantifies functional information in the identified tissue of interest in the functional image data.

4. The image data processor of claim 1, wherein the multi-structure atlas includes a mapping that physically maps locations of sub-structure of the structure of interest to the tissue of interest for each sub-image data of the image data corresponding to at least one of a different examination or a different subject, and the structural image data processor segments the region of interest based on the sub-structure in the atlas so that the region of interest includes the tissue of interest.

5. The image data processor of claim 4, wherein the mapping includes at least one of 1) a point cloud of the tissue of interest on or within an organ surface;

2) a 3D sheet having the tissue of interest across an organ;

     3) or a 3D structure in a form of region of interest.

6. The image data processor of claim 4, further comprising:

a multi-structure atlas generator that generates the multi-structure atlas, the multi-structure atlas generator, comprising;

a structure of interest segmentor that automatically segments the structure from the structural image data based on a predetermined algorithm;

a tissue of interest segmentor that segments the tissue of interest from the structural image data based on user input;

a structure to tissue mapper that generates a map mapping a physical spatial location of each sub-structure to the segmented tissue of interest; and

a modeler that generates the multi-structure atlas based on a plurality of the maps.

18. The image data processor of claim 1, wherein, in response to a lung lobe boundary being located between a first rib and a second different rib in the tissue of interest, the functional image data processor identifies the region of interest from a location of the first and second ribs in the multi-structure map, and the region of interest includes the lung lobe boundary.

19. The image data processor of claim 18, wherein the functional image data processor localizes the lung lobe boundary based on the first and second ribs.

20. The image data processor of claim 18, wherein the functional image data processor determines the region of interest includes a region around the first and second ribs.

21. The image data processor of claim 18, wherein the functional image data processor determines the region of interest includes a region between the first and second ribs.

22. The image data processor of claim 18, wherein the multi-structure map indicates a lung lobe boundary is located a first distance from the first rib and a second distance from the second rib.

23. The image data processor of claim 18, wherein the multi-structure map includes a numerical value that indicates a physical spacing of the lung lobe boundary from a rib on a surface of the lung.

24. The image data processor of claim 1, wherein the multi-structure map indicates a thickness of the region of interest, in a form of the 3D sheet, with fissures across a lung volume.

25. The image data processor of claim 1, wherein the point cloud includes a set of 3D points on the lung boundary.

7. An apparatus, comprising:

a processor that generates a multi-structure atlas, wherein the multi-structure atlas physically maps structure of interest to tissue of interest such that locating the structure of interest in structural image data based on the multi-structure atlas localizes the tissue of interest to the region of interest, wherein the processor generates the multi-structure atlas by;

automatically segmenting the structure from the structural image data based on a predetermined algorithm;

segmenting the tissue of interest from the structural image data with user interaction;

mapping a physical spatial location of the segmented structure to the segment tissue of interest; and

generating the multi-structure atlas based on a plurality of maps, wherein each map corresponds to sub-set of the image data for at least one of a different examination or a different subject.

8. The apparatus of claim 7, the processor is further configured to:

employ the multi-structure atlas to segment a region of interest from structural image data under evaluation which includes the tissue of interest and that segments the tissue of interests from the region of interest; and

register the structural image data and the functional image data and that identifies the tissue of interest in the functional image data based on the segmented tissue of interest and the registered image data.

9. The apparatus of claim 8, the processor further configured to:

quantify functional information in the identified tissue of interest.

10. A method, comprising:

employing a multi-structure atlas to segment a region of interest from structural image data which includes tissue of interest and that segments the tissue of interests from the region of interest, wherein the structure is rib cage and the tissue of interest is lung lobe, and the multi-structure atlas is a rib cage/lung lobe atlas that includes a mapping that maps one or more ribs to a location of a boundary of a lung lobe, and the mapping includes a lung lobe region of interest, in a form of a least one of a point cloud on a surface of the lung or a 3D sheet across lung, through the rib case/lung lobe atlas; and

identifying the tissue of interest in functional image data based on the segmented tissue of interest.

11. The method of claim 10, further comprising:

registering the structural image data and the functional image data and identifying the tissue of interest in the functional image data based on the segmented tissue of interest and the registered image data.

12. The method of claim 10, further comprising:

quantifying functional information in the identified tissue of interest.

13. The method of claim 10, wherein the multi-structure atlas physically maps structure to the tissue of interest for each sub-image data of the image data corresponding to at least one of a different examination or a different subject, and the structural image data processor segments the region of interest based on the structure in the atlas so that the region of interest includes the tissue of interest.

14. The method of claim 13, further comprising:

automatically segmenting the structure from the structural image data based on a predetermined algorithm;

segmenting the tissue of interest segmentor from the structural image data under user interaction;

mapping a physical spatial location of the segmented structure to the segment tissue of interest; and

generating the multi-structure atlas based on a plurality of mappings, each corresponding to a different sub-set of the image data.

15. A method, comprising:

generating a multi-structure atlas, wherein the multi-structure atlas physically maps structure to tissue of interest such that locating the structure in structural image data based on the multi-structure atlas localizes the tissue of interest to the region of interest, wherein generating the multi-structure atlas includes;

automatically segmenting the structure from the structural image data based on a predetermined algorithm;

segmenting the tissue of interest segmentor from the structural image data under user interaction;

mapping a physical spatial location of the segmented structure to the segment tissue of interest; and

generating the multi-structure atlas based on a plurality of mappings, each corresponding to a different sub-set of the image data.

16. The method of claim 15, further comprising:

employing the multi-structure atlas to segment a region of interest from structural image data that includes tissue of interest and that segments the tissue of interests from the region of interest; and

registering the structural image data and the functional image data; and

identifying the tissue of interest in the functional image data based on the segmented tissue of interest and the registered image data.

17. The method of claim 16, further comprising:

quantifying functional information in the identified tissue of interest.