Methods and systems for 3D segmentation of ultrasound images

• US 7,428,334 B2
• Filed: 08/27/2004
• Issued: 09/23/2008
• Est. Priority Date: 08/27/2004
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Schoisswohl, Armin, Ruisz, Johannes
• Primary Examiner(s): Mehta; Bhavesh
• Assistant Examiner(s): Strege; John B
• Application Number:
  US10/927,827
• Publication Number:
  US 20060056690A1
• Time in Patent Office: 1,488 Days
• Field of Search: 382/128, 382/173, 382/154, 382/199
• US Class Current: 382/173
• CPC Class Codes: G06T 2207/10136 3D ultrasound image
• CPC Class Codes: G06T 2207/20168 Radial search
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/12 Edge-based segmentation

Abstract

A method for three dimensional (3D) segmentation of an object is provided. The method obtains a volumetric data set containing object data and non-object data in proximity to the object data, the object data having a reference axis extending through the object data. The method defines at least one reference slice and multiple object slices within the volumetric data set, the reference slice and the object slices intersecting one another along the reference axis and containing the reference axis. Reference points are determined within the reference slice at edges of the object data. With reference points determined, the method generates an estimated contour extended through the reference and object slices based on the reference points, the estimated contour intersecting the object slices to define estimated contour points. The method then adjusts the estimated contour points until corresponding substantially to actual contour points of the object data.

Claims

1. A method for three dimensional (3D) segmentation of an object, comprising:

obtain a volumetric data set containing object data and non-object data in proximity to said object data, said object data having a reference axis extending through said object data;

defining at least one reference slice and multiple object slices within said volumetric data set, said reference slice and said object slices intersecting one another along said reference axis and containing said reference axis;

determining, within said reference slice, reference points at edges of said object data;

generating an estimated contour extended through said reference and object slices based on said reference points, said estimated contour intersecting said object slices to define estimated contour points; and

adjusting said estimated contour points until corresponding substantially to actual contour points of the object data.

2. The method of claim 1, further comprising, adjusting said estimated contour point within a first object slice based on a difference between estimated and actual contour points in a second object slice.

3. The method of claim 1, wherein said obtaining includes obtaining a volumetric data set containing object data, non-object data surrounding said object data, said object data having said reference axis located at and extending through a center of said object data.

4. The method of claim 1, wherein said defining includes defining orthogonal reference slices and said determining includes determining pairs of reference points in each of said orthogonal reference slices.

5. The method of claim 1, wherein said estimated contour is substantially a circle.

6. The method of claim 1, further comprising repeating said defining, determining, generating and adjusting at multiple depths within said reference and object slices, where depth is in a direction extending parallel to said reference axis.

7. The method of claim 1, further comprising establishing limits for an amount that estimated contour points are adjusted in one operation.

8. The method of claim 1, wherein said estimated contour points are adjusted in an iterative manner until corresponding substantially to said actual contour points.

9. The method of claim 1, further comprising determining a direction in which to adjust said estimated contour point based on a line normal to said estimated curve.

10. The method of claim 1, further comprising, for each object slice, determining a first distance to adjust said estimated contour point and reducing said first distance based on object data and a smoothness of said estimated contour.

11. The method of claim 1, further comprising reducing an amount to adjust said estimated contour point based on at least one of a gradient characteristic, local statistics, texture measurements, and a smoothness of said estimated contour.

12. The method of claim 1, wherein said adjusting is performed in an iterative manner until a predetermined percentage of said estimated contour points are not moved during an adjustment iteration.

13. A system for three dimensional (3D) segmentation of an object, comprising:

memory storing a volumetric data set containing object data and non-object data in proximity to said object data, said object data having a reference axis extending through said object data;

a processing unit defining at least one reference slice and multiple object slices within said volumetric data set, said reference slice and said object slices intersecting one another along said reference axis and containing said reference axis, said processing unit determining, within said reference slice, reference points at edges of said object data;

a contour estimator generating an estimated contour extended through said reference and object slices based on said reference points, said estimated contour intersecting said object slices to define estimated contour points in each object slice;

a contour adjustment unit adjusting said estimated contour points until corresponding substantially to actual contour points of the object data.

14. The system of claim 13, wherein said contour adjustment unit adjusts said estimated contour point within a first object slice based on a difference between estimated and actual contour points in a second object slice.

15. The system of claim 13, wherein said processing unit defines orthogonal reference slices and determines a pair of reference points in each of said orthogonal reference slices.

16. The system of claim 13, wherein said processing unit, contour estimator and contour adjustment unit repeat said defining, determining, generating and adjusting operations at multiple depths within said reference and object slices, where depth is in a direction extending parallel to said reference axis.

17. The system of claim 13, wherein said contour adjustment unit establishes limits for an amount that estimated contour points are adjusted in one operation.

18. The system of claim 13, wherein said estimated contour points are adjusted in an iterative manner until corresponding substantially to said actual contour points.

19. The system of claim 13, wherein said contour adjustment unit determines a direction in which to adjust said estimated contour point based on a line normal to said estimated curve.

20. The system of claim 13, wherein said contour adjustment unit determines, for each object slice, a first distance to adjust said estimated contour point and reduces said first distance based on object data and a smoothness of said estimated contour.

21. The system of claim 13, wherein said contour adjustment unit reduces an amount to adjust said estimated contour point based on at least one of a gradient characteristic, local statistics, texture measurements, and a smoothness of said estimated contour.

22. The system of claim 13, wherein said contour adjustment unit adjusts said estimated contour points in an iterative manner until a predetermined percentage of said estimated contour points are not moved during an adjustment iteration.