Automated identification of an anatomy part

Patent metadata

Field	Value
Publication number	US-10524741-B2
Application number	US-201113637373-A
Country	US
Kind code	B2
Filing date	Mar 24, 2011
Priority date	Mar 31, 2010
Publication date	Jan 7, 2020
Grant date	Jan 7, 2020

Abstract

The present invention relates to a device 10 for automatically identifying a part 20a, 20b of an anatomy structure comprising several parts 20a, 20b, in which anatomy structure an intervention device 21 resides. The device 10 comprises a feature extraction unit 11 and an anatomy part classification unit 13. The feature extraction unit 11 uses provided image content data ICD to extract at least one characterizing feature DS of the appearance of 10 the intervention device 21. The anatomy part classification unit 13 correlates the at least one characterizing feature DS with provided classifier data CD which are characteristic for a projection feature of the intervention device 21 viewed under certain geometry of an imaging system 30. After correlating, the anatomy part classification unit 13 determines in which part 20a, 20b of the anatomy structure comprising several parts 20a, 20b the intervention device 2115 is located.

First claim

The invention claimed is:

1. A system configured for automatically identifying an anatomy part of an anatomy structure comprising two or more anatomy parts, in which anatomy structure an intervention device resides, the system comprising:
an imaging system for imaging the anatomy structure using radiation and deriving image content data from the imaging; and
a processor programmed to:
extract at least one characterizing feature of an appearance of the intervention device using the derived image content data;
correlate the at least one characterizing feature with provided classifier data that is linked correspondingly to particular anatomical categories into which said parts would be placeable in classifying said parts; and
determine, based on said linked anatomical categories, in which part, from among said two or more anatomy parts, the intervention device is located during an interventional procedure.

2. The system according to claim 1 , wherein the processor is further programmed to estimate projection characteristics of the appearance of the intervention device being located in a part from among said two or more anatomy parts of the anatomy structure using provided system geometry data and provided three dimensional data of a model of the intervention device being located in a part from among said two or more anatomy parts of the anatomy structure;
wherein the estimated projection characteristics of the appearance of the intervention device are used to extract the at least one characterizing feature.

3. The system according to claim 1 , wherein the intervention device is a catheter, and the part from among said two or more anatomy parts of the anatomy structure is a left coronary ostium or a right coronary ostium of an aorta.

4. The system according to claim 1 , further comprising:
a display unit connected with the processor for displaying at least a portion of the image content data.

5. The system of claim 1 , wherein said categories include heart structures comprising a left atrium, a right atrium, a left ventricle and a right ventricle.

6. The system of claim 1 , wherein said categories include an intervention side of said anatomy structure.

7. The system of claim 1 , said two or more parts amounting to more than two.

8. The system of claim 1 , said determining entailing selecting from among alternative, said alternative respectively representing ones of said parts.

9. A system for automatically identifying a part of an anatomy structure comprising two or more parts, in which anatomy structure an intervention device resides, the system comprising:
an imaging system for imaging the anatomy structure and deriving image content data from the imaging; and
a processor programmed to:
extract at least one characterizing feature of an appearance of the intervention device using the derived image content data;
correlate the at least one characterizing feature with provided classifier data and to determine in which part, from among the two or more parts, the intervention device is located during an interventional procedure; and
generate classifier data using provided three dimensional data of a model of the intervention device being located in a part of the two or more parts of the anatomy structure and using provided system geometry data of the imaging system.

10. The system of claim 9 , said model linking a hypothetical predetermined location of said intervention device with respect to said parts to a respective observable shape of said intervention device.

11. The system of claim 9 , said model linking a hypothetical predetermined location of said intervention device with respect to said parts to a respective observable radio-opaqueness of said device.

12. The system of claim 9 , said two or more parts amounting to more than two.

13. A method for automatically identifying a part of an anatomy structure comprising two or more parts, in which anatomy structure an intervention device resides during an interventional procedure, the method comprising:
imaging the anatomy structure, including an appearance of the intervention device residing in the anatomy structure, to provide image content data;
extracting a characterizing feature of the appearance of the intervention device using the provided image content data;
correlating the extracted characterizing feature with provided classifier data;
determining in which part of the two or more parts of the anatomy structure the intervention device is located during the interventional procedure; and
providing the classifier data using provided three dimensional data of a model of the intervention device being located in a part of the two or more parts of the anatomy structure and using provided system geometry data of an imaging system.

14. The method according to claim 13 , further comprising estimating projection characteristics of the intervention device being located in a part of the two or more parts of the anatomy structure using provided system geometry data and provided three dimensional data of a model of the intervention device being located in a part of the two or more parts of the anatomy structure; wherein projection characteristics of the appearance of the intervention device are estimated to produce estimated projection characteristics that are used to perform the extracting of the characterizing feature of the intervention device.

15. The method according to claim 14 , further comprising deriving at least one parameter from the extracted characterizing feature of the intervention device located in a part of the two or more parts of the anatomy structure; wherein the classifier data is at least one classifier parameter characteristic for a projection feature of the intervention device located in a part of the two or more parts of the anatomy structure.

16. The method according to claim 15 , wherein in deriving the at least one parameter, a three dimensional model of the aortic cross, the aortic root, the right coronary ostium and the left coronary ostium are used to generate the at least one classifier parameter.

17. The method according to claim 13 , wherein it is determined that the intervention device is located in a part of the two or more parts of the anatomy structure when the extracted characterizing feature is in a predetermined range of the classifier data.

18. The method according to claim 13 , wherein the intervention device is a catheter; and the part of the two or more parts of the anatomy structure is a left coronary ostium or a right coronary ostium of an aorta.

19. The method of claim 13 , said two or more parts amounting to more than two.

20. A non-transitory computer readable medium for automatically identifying an anatomy part of an anatomy structure comprising two or more anatomy parts, in which anatomy structure an intervention device resides during an interventional procedure, said medium embodying a program having instructions executable by a processor for performing a method comprising:
causing said anatomy structure to be interrogated using radiation;
based on a result of the interrogating, deriving image content data;
extracting a characterizing feature from the derived image content data of an appearance of the intervention device;
correlating the extracted characterizing feature with provided classifier data; and
determining in which anatomy part of the two or more anatomy parts of the anatomy structure the intervention device is located during the interventional procedure,
said classifier data comprising a characte

Assignees

• Florent Raoul
• Den Hartog Willem Frederik
• Auvray Vincent Maurice Andre
• Koninklijke Philips Nv

Inventors

• Florent Raoul
• Den Hartog Willem Frederik
• Auvray Vincent Maurice Andre

Classifications

• G16H50/30

  for calculating health indices; for individual health risk assessment · CPC title

• A61B8/0891

  for diagnosis of blood vessels · CPC title

• G06T2207/10121

  Fluoroscopy · CPC title

• A61B8/0833

  involving detecting or locating foreign bodies or organic structures · CPC title

• A61B6/504

  for diagnosis of blood vessels, e.g. by angiography · CPC title