Linking breast lesion locations across imaging studies

Patent metadata

Field	Value
Publication number	US-10109048-B2
Application number	US-201414893679-A
Country	US
Kind code	B2
Filing date	Jun 20, 2014
Priority date	Jun 28, 2013
Publication date	Oct 23, 2018
Grant date	Oct 23, 2018

Abstract

The present invention provides for means for linking breast lesion locations across imaging studies. In particular, a generic three-dimensional representation of the female breast is used. Automatic translation of the lesion location into standard clinical terminology and aligning the breast model with individual patient images is comprised. Moreover, a mechanism for linking image locations showing a lesion to a location in the breast model is presented. If desired, a region of interest can be calculated by a region of interest definition module that predicts a region of interest of a known lesion in terms of the breast model representation in a new imaging study.

First claim

The invention claimed is:

1. A breast model based method for linking breast lesion locations, the method comprising:
receiving first volumetric image data of a first image of a patient's breast,
providing coordinates of a lesion location in the first image of the patient's breast;
providing a predefined generic three-dimensional (3D) geometric breast model that is a representation of a female breast;
wherein the predefined generic 3D geometric breast model is based on a normalized 3D breast coordinate system;
mapping the first volumetric image data with the predefined generic 3D geometric breast model; and
linking the lesion location in the first image with a corresponding location in the predefined generic 3D geometric breast model thereby providing linking data;
receiving second volumetric image data of a second image of a patient's breast;
mapping the second image data with the predefined generic 3D geometric breast model;
automatically determining a location of the lesion in the second image based on the linking data,
automatically locating a region of interest of the second volumetric image data corresponding to the determined location of the lesion in the second image, and
automatically displaying one or more images corresponding to the located region of interest of the second volumetric image data.

2. The method of claim 1 , wherein the breast model comprises as anatomic landmarks for mapping at least one element chosen from the group comprising: skin progression, breast nipple, pectoral muscle, anatomic orientation, infra-mammary fold, and any combination thereof, and wherein the mapping comprises aligning at least one anatomic landmark of the breast model with a corresponding anatomic landmark in the first image.

3. The method of claim 1 , further comprising applying at least one detection algorithm to the first image thereby detecting at least one anatomic landmark in the first image.

4. The method of claim 3 , further comprising shifting and/or rotating the predefined generic geometric breast model such that a part of the model that corresponds to the detected at least one anatomic landmark of the first image is aligned with the at least one detected anatomic landmark in the first image.

5. The method of claim 1 , further comprising scaling either the breast model or the first image such that the size of the breast model corresponds to the size of the breast of the first image.

6. The method of claim 1 , wherein the breast model comprises an elastic mesh structure, the method further comprising shifting inner mesh vertices to skin points in the first image to minimize a mesh deformation energy.

7. The method of claim 1 , the method further comprising automatically performing segmentation or image analysis of the second image data at or around the determined lesion location in the second image.

8. The method of claim 1 , wherein the mapping of the second image data includes applying image analysis to the second image of a patient's breast thereby detecting at least one anatomic landmark of the breast model in the second image of the patient's breast.

9. The method of claim 1 , wherein the first image is of a first imaging modality and the second image is of a second, different imaging modality.

10. An imaging system for imaging a patient's breast, wherein the imaging system is configured to execute the method of claim 1 .

11. A non-transitory computer readable medium that comprises a computer program element for linking breast lesion locations which program element, when being executed by a processor, causes the processor to:
receive first volumetric image data of a first image of a patient's breast,
receive coordinates of a lesion location in the first image of the patient's breast;
provide a predefined generic three-dimensional (3D) geometric breast model that is a representation of a female breast;
wherein the predefined generic 3D geometric breast model is based on a normalized breast coordinate system;
map the first image data with the predefined generic geometric breast model;
link the lesion location of the lesion in the first image with a corresponding location in the breast model thereby providing linking data,
receive second image data of a second image of a patient's breast,
map the second image data with the predefined generic geometric breast model, and
determine a location of the lesion in the second image based on the stored linking data,
locate a region of interest of the second volumetric image data corresponding to the determined location of the lesion in the second image, and
display one or more images corresponding to the located region of interest of the second volumetric image data.

12. The medium of claim 11 , wherein the breast model comprises as anatomic landmarks for mapping at least one element chosen from the group comprising: skin progression, breast nipple, pectoral muscle, anatomic orientation, infra-mammary fold, and any combination thereof, and wherein the mapping comprises aligning at least one anatomic landmark of the breast model with a corresponding anatomic landmark in the first image.

13. The medium of claim 11 , wherein the program element causes the processor to apply at least one detection algorithm to the first image to detect at least one anatomic landmark in the first image.

14. The medium of claim 13 , wherein the program element causes the processor to shift and/or rotate the predefined generic geometric breast model such that a part of the model that corresponds to the detected at least one anatomic landmark of the first image is aligned with the at least one detected anatomic landmark in the first image.

15. The medium of claim 11 , wherein the program element causes the processor to scale either the breast model or the first image such that the size of the breast model corresponds to the size of the breast of the first image.

16. The medium of claim 11 , wherein the breast model comprises an elastic mesh structure, and the program element causes the processor to shift inner mesh vertices to skin points in the first image to minimize a mesh deformation energy.

17. The medium of claim 11 , wherein the program element causes the processor to perform segmentation or image analysis in the second image data at or around the determined lesion location in the second image.

18. The medium of claim 11 , wherein the program element causes the processor to apply image analysis to the second image of a patient's breast to detect at least one anatomic landmark of the breast model in the second image of the patient's breast.

19. The medium of claim 11 , wherein the first image is of a first imaging modality and the second image is of a second, different imaging modality.

20. The medium of claim 11 , wherein the program element causes the processor to concurrently display images of the region of interest in the first image and in the second image.

Assignees

• Koninklijke Philips Nv

Inventors

• Buelow Thomas
• Bergtholdt Martin
• Wiemker Rafael
• Meetz Kirsten Regina
• Sabczynski Joerg
• Kutra Dominik Berjamin

Classifications

• G06T2207/10116

  X-ray image · CPC title

• G06T2207/30096

  Tumor; Lesion · CPC title

• G06T7/33 Primary

  using feature-based methods · CPC title

• G06T2207/10132

  Ultrasound image · CPC title

• G06T2207/30068

  Mammography; Breast · CPC title