System and method for pose estimation of an imaging device and for determining the location of a medical device with respect to a target

Patent metadata

Field	Value
Publication number	US-11712213-B2
Application number	US-202117142642-A
Country	US
Kind code	B2
Filing date	Jan 6, 2021
Priority date	Feb 8, 2018
Publication date	Aug 1, 2023
Grant date	Aug 1, 2023

Abstract

A system and method for estimating a pose of an imaging device for one or more images is provided.

First claim

What is claimed is:

1. A method for estimating a pose of an imaging device, the method comprising:
generating a probability map based on a captured image of a structure of markers captured by the imaging device;
generating candidates for a possible image of the structure of markers captured by the imaging device corresponding to possible poses of the imaging device, yielding generated candidates;
identifying a candidate from the generated candidates having the highest probability of being the captured image of the structure of markers based on the probability map, yielding an identified candidate; and
estimating the pose of the imaging device for the captured image based on a possible pose of the imaging device corresponding to the identified candidate, yielding an estimated pose.

2. The method of claim 1 , further comprising:
generating a locally deformed version of the identified candidate to maximize its probability of being the captured image of the structure of markers based on the probability map;
determining an improved candidate for the captured image of the structure of markers based on the locally deformed version of the identified candidate; and
determining the possible pose of the imaging device associated with the improved candidate as the estimated pose of the imaging device for the captured image.

3. The method of claim 2 , further comprising repeating: generating a locally deformed version of the identified candidate and determining an improved candidate.

4. The method of claim 1 , wherein generating a probability map includes generating an image of smooth densities based on the captured image.

5. The method of claim 4 , wherein the image of smooth densities includes pixels indicating a probability that the pixels show a portion of the structure of markers.

6. The method of claim 4 , wherein the image of smooth densities is generated using a Harris corner detector.

7. The method of claim 1 , wherein markers of the structure of markers are arranged in a periodic pattern.

8. The method of claim 1 , further comprising:
capturing a sequence of images;
determining non-sequential images of the sequence of images;
estimating the pose of the imaging device for the non-sequential images;
reducing a range of possible poses for the imaging device based on the estimated poses of the imaging device for the non-sequential images, yielding a reduced range; and
estimating the pose of the imaging device for images of the sequence of images other than the non-sequential images according to the reduced range.

9. A system for estimating a pose of an imaging device for an image acquired by the imaging device, the system comprising:
a structure of markers;
an imaging device configured to acquire an image of at least a portion of the structure of markers;
at least one processor; and
a memory having stored thereon instructions, which, when executed by the at least one processor, causes the at least one processor to:
generate a probability map based on a captured image of a structure of markers captured by the imaging device;
generate candidates for possible images of the structure of markers captured by the imaging device at different possible poses, yielding generated candidates; and
identify a candidate from the generated candidates having the highest probability of being the captured image of the structure of markers based on the probability map, yielding an identified candidate.

10. The system of claim 9 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
generate a locally deformed version of the identified candidate to locally maximize the identified candidate's probability of being the captured image of the structure of markers based on the probability map; and
determine an improved candidate for the captured image of the structure of markers based on the locally deformed version of the identified candidate.

11. The system of claim 9 , wherein each marker of the structure of markers is a radio-opaque sphere.

12. The system of claim 9 , wherein markers of the structure of markers are arranged in aurid pattern.

13. The system of claim 9 , wherein the imaging device is a fluoroscopic imaging device.

14. The system of claim 9 , wherein generating a probability map includes generating an image of smooth densities using a marker detector.

15. A system for estimating poses for a sequence of images, the system comprising:
at least one processor; and
a memory having stored thereon instructions, which, when executed by the at least one processor, causes the at least one processor to:
generate probability maps based on a sequence of images of a structure of markers;
generate, for each image of the sequence of images, candidates for possible images of the structure of markers at different possible poses of an imaging device, yielding generated candidates; and
identify, for each image of the sequence of images, a candidate from the generated candidates having the highest probability of being the image of the structure of markers based on the probability map for the image, yielding identified candidates.

16. The system of claim 15 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
generate locally deformed versions of the identified candidates to locally maximize their probability of being the sequence of images of the structure of markers based on the probability maps; and
determine improved candidates for possible images of the structure of markers based on the locally deformed versions of the identified candidates.

17. The system of claim 15 , wherein markers of the structure of markers are radio-opaque spheres.

18. The system of claim 15 , wherein markers of the structure of markers are arranged in a grid pattern.

19. The system of claim 15 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
determine non-sequential images of the sequence of images;
estimate poses for the non-sequential images, yielding estimated poses for the non-sequential images;
reduce a range of the different possible poses based on the estimated poses for the non-sequential images, yielding a reduced range; and
estimate poses for images of the sequence of images other than the non-sequential images according to the reduced range.

20. The system of claim 15 , wherein markers of the structure of markers are arranged in a two-dimensional pattern.

Assignees

• Covidien Lp

Inventors

• Barak Ron
• Birenbaum Ariel
• Alexandroni Guy
• Weingarten Oren P

Classifications

• A61B6/487 Primary

  involving fluoroscopy · CPC title

• A61B1/00158

  using magnetic field · CPC title

• A61B1/043

  for fluorescence imaging · CPC title

• A61B5/055

  involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging · CPC title

• A61B5/064

  using markers (A61B5/062 takes precedence) · CPC title