Systems and methods for automated distortion correction and/or co-registration of three-dimensional images using artificial landmarks along bones

Patent metadata

Field	Value
Publication number	US-11182913-B2
Application number	US-201816615618-A
Country	US
Kind code	B2
Filing date	Jun 15, 2018
Priority date	Jun 16, 2017
Publication date	Nov 23, 2021
Grant date	Nov 23, 2021

Abstract

Presented herein are systems and methods for registering one or more images of one or more subjects based on the automated generation of artificial landmarks. An artificial landmark is a point within an image that is associated with a specific physical location of the imaged region. The artificial landmarks are generated in an automated and robust fashion along the bones of a subject's skeleton that are represented in the image (e.g. graphically). The automatically generated artificial landmarks are used to correct distortion in a single image or to correct distortion in and/or co-register multiple images of a series of images (e.g. recorded at different time points). The artificial landmark generation approach described herein thereby facilitates analysis of images used, for example, for monitoring the progression of diseases such as pulmonary diseases.

First claim

What is claimed is:

1. A system for registration of one or more 3-D images of a subject, the system comprising:
a processor; and
a memory with instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to:
(a) receive a 3-D image of the subject, wherein the 3-D image comprises a graphical representation of one or more bones of interest;
(b) identify in the graphical representation one or more bones of interest and a reference object;
(c) automatically generate one or more seed artificial objects, each seed artificial object being within a first distance interval from the reference object and corresponding to a region of a bone of interest in the graphical representation;
(d) for each automatically generated seed artificial object, automatically generate a plurality of associated subsequent artificial objects, thereby creating a set of artificial objects within the image;
(e) automatically perform registration of one or more images of the subject using the set of artificial objects within the image;
(f) identify one or more dangerous regions of the image corresponding to regions in which a distance between a first and second bone of interest in the graphical representation is below a predefined threshold distance; and
(g) automatically generate each artificial object such that it is sufficiently far from any identified dangerous region within the image.

2. The system of claim 1 , wherein the instructions cause the processor to automatically generating the one or more seed artificial objects by:
determining a distance map comprising intensity values at each of a plurality of points in three dimensions, each of the intensity values of the distance map corresponding to a distance from a given point in 3-D space to the reference object;
generating a distance interval mask from the distance map; and
applying an AND operation between the distance interval mask and a mask corresponding to the identified one or more bones of interest, thereby identifying a plurality of regions of the identified bones of interest that are within the first distance interval from the reference object.

3. The system of claim 1 , wherein for each automatically generated seed artificial object, the plurality of associated subsequent artificial objects comprises an equidistant chain of artificial objects.

4. The system of claim 1 , wherein each artificial object is a predefined threshold distance from an identified additional bone within the graphical representation.

5. The system of claim 1 , wherein the instructions cause the processor to automatically generate the plurality of subsequent artificial objects associated with a respective seed artificial object by generating a chain of artificial objects along a bone of interest by beginning with the seed artificial object and, in a stepwise fashion, generating new subsequent artificial objects, each newly generated artificial object proceeding outwards, away from the identified reference object, along the bone of interest.

6. The system of claim 5 , wherein the instructions cause the processor to generate the chain of artificial objects associated with the respective seed artificial object by, for at least one newly generated artificial object of the chain of artificial objects:
determining whether the newly generated artificial object is within a predetermined threshold distance from an identified additional bone in the graphical representation; and
responsive to determining that the newly generated artificial object is within the predetermined threshold distance from the identified additional bone of the image, terminating generation of subsequent artificial objects associated with the respective seed artificial object.

7. The system of claim 1 , wherein each artificial object of the set of artificial objects within the image is confirmed to have at least a predefined threshold volume.

8. The system of claim 1 , wherein the instructions cause the processor to:
determine, based on the set of artificial objects within the image, an image registration transformation; and
apply the image registration transformation to a region of the 3-D image, thereby registering the 3-D image.

9. The system of claim 8 , wherein the image registration transformation yields symmetrization of the 3-D image, thereby correcting distortions in the image.

10. The system of claim 8 , wherein the received image corresponds to a first image, and the image registration transformation aligns the first image with a second image of the subject, thereby co-registering the first image with the second image.

11. The system of claim 8 , wherein the instructions cause the processor to determine the image registration transformation by:
determining, from the set of artificial objects within the image, a set of artificial landmarks within the image, each landmark corresponding to a point determined from a corresponding artificial object; and
determining the image registration transformation using the set of artificial landmarks within the image and a set of target landmarks, wherein the image registration transformation is determined to, when applied to points corresponding to the artificial landmarks within the image, substantially optimize alignment of the set of artificial landmarks with the set of target landmarks.

12. The system of claim 11 , wherein the set of target landmarks is symmetric.

13. The system of claim 11 , wherein the instructions cause the processor to determine the set of target landmarks using the set of artificial landmarks within the image.

14. The system of claim 11 , wherein the set of target landmarks is a set of predetermined target landmarks.

15. The system of claim 8 , wherein the received 3-D image comprises one or more regions corresponding to graphical representations of soft tissue, and the instructions cause the processor to apply the image registration transformation to the one or more regions of the image corresponding to graphical representations of soft tissue, thereby registering the soft tissue regions.

16. The system of claim 8 , wherein the received 3-D image of the subject corresponds to a first image recorded via a first modality, and, wherein the instructions cause the processor to:
receive a second image recorded via a second modality;
determine, based on the set of artificial objects within the image, a first image registration transformation;
determine, based on the first image registration transform, a second image registration transformation; and
apply the second image registration transformation to a region of the second image.

17. The system of claim 16 , wherein the second image is recorded at substantially the same time as the first image, with the subject in a substantially similar pose and/or position.

18. The system of claim 16 , wherein the second image registration transformation is the same as the first image registration transformation.

19. The system of claim 16 , wherein coordinates of a plurality of points of the first image are related to coordinates of a plurality of points of the second image via a known functional relationship.

20. The system of claim 1 , wherein:
the 3-D image comprises a graphical representation of rib bones and a backbone,
the identified one or more bones of interest comprise rib bones, and
the identified reference object is a backbone of the subject.

21. The system of claim 20 , wherein:
the graphical representation of rib bones includes a plurality

Assignees

• Perkinelmer Cellular Tech Germany Gmbh
• Perkinelmer Health Sci Inc

Inventors

• Kask Peet
• Meganck Jeffrey

Classifications

• G06T7/32 Primary

  using correlation-based methods · CPC title

• G16H30/40 Primary

  for processing medical images, e.g. editing · CPC title

• G06T7/30

  Determination of transform parameters for the alignment of images, i.e. image registration · CPC title

• G16H50/50

  for simulation or modelling of medical disorders · CPC title

• G16H30/20

  for handling medical images, e.g. DICOM, HL7 or PACS · CPC title