System and Method for Detecting and Matching Anatomical Structures Using Appearance and Shape

• US 20090226081A1
• Filed: 03/11/2009
• Published: 09/10/2009
• Est. Priority Date: 11/19/2003
• Status: Active Grant

• Current Assignee: Siemens Healthcare GMBH (Siemens AG)
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG), Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Gupta, Alok, Georgescu, Bogdan, Zhou, Xiang Sean, Rao, R. Bharat, Comaniciu, Dorin
• Granted Patent: US 7,747,054 B2
• Time in Patent Office: Days
• US Class Current: 382/159
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/30048 Heart; Cardiac
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06V 10/754 involving a deformation of the sample pattern or of the reference pattern; Elastic matching
• CPC Class Codes: G06V 10/774 Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting

Abstract

A detection framework that matches anatomical structures using appearance and shape is disclosed. A training set of images are used in which object shapes or structures are annotated in the images. A second training set of images represents negative examples for such shapes and structures, i.e., images containing no such objects or structures. A classification algorithm trained on the training sets is used to detect a structure at its location. The structure is matched to a counterpart in the training set that can provide details about the structure'"'"'s shape and appearance.

Claims

1. A method for detecting and matching anatomical structures in an image to one or more anatomical structures in a training set of images comprising the steps of:

receiving a candidate image;

extracting feature values from the candidate image;

applying a classification function to detect an anatomical structure;

if an anatomical structure is detected, identifying one or more counterpart images in the training set of images by matching the extracted feature values for the candidate image to feature values of the counterpart images in the training set; and

using one or more shapes of anatomical structures in the matching counterpart images from the training set to determine a shape of the anatomical structure in the candidate image.

2. The method of claim 1 wherein the step of matching feature values further comprises the steps of:

generating a feature vector for the candidate anatomical structure; and

comparing the candidate feature vector to a feature vector associated with a counterpart image.

3. The method of claim 2 wherein the feature vector comprises a plurality of weak classifier outputs h_i'"'"'s and associated weights a_i'"'"'s.

4. The method of claim 3 wherein the matching step further comprises using different distance measures, including using a weighting matrix in the Euclidean distance function.

5. The method of claim 1 wherein said step of applying a classification function further comprises the step of:

scanning a window through the candidate image.

6. The method of claim 1 wherein said applying step is performed in a hierarchical manner.

7. The method of claim 1 wherein the classification function uses boosting.

8. The method of claim 1 wherein the classification function uses AdaBoosting.

9. The method of claim 1 wherein the classification function uses RankBoosting.

10. The method of claim 1 wherein the classification function uses a support vector machine.

11. The method of claim 1 wherein the image is an ultrasound image.

12. The method of claim 11 wherein the anatomical structure is a left ventricle.

13. The method of claim 1 wherein the anatomical structure is a face.

14. A method for matching an anatomical structure in an image to one or more similarly shaped anatomical structures in a training set of images comprising the steps of:

receiving an image of a candidate anatomical structure;

extracting features from the image;

comparing features associated with similarly shaped anatomical structures to the candidate anatomical structure; and

determining the shape of the candidate anatomical structure by using the shape of at least one nearest neighbor from the training set.

15. The method of claim 14 wherein the candidate anatomical structure has its contour annotated by a set of control points.

16. The method of claim 14 wherein each of the similarly shaped anatomical structures has its contour annotated by a set of control points.

17. The method of claim 14 wherein the step of comparing features further comprises the steps of:

generating a feature vector for the candidate anatomical structure; and

comparing the candidate feature vector to a feature vector associated with a similarly shaped anatomical structure.

18. The method of claim 17 wherein the feature vector comprises a plurality of weak classifier outputs h_i'"'"'s and associated weights a_i'"'"'s.

19. The method of claim 18 wherein the comparing step further comprises using different distance measures, including using a weighting matrix in the Euclidean distance function.

20. The method of claim 14 wherein the image is an ultrasound image.

21. The method of claim 20 wherein the anatomical structure is a left ventricle.

22. The method of claim 14 wherein the anatomical structure is a face.

23. A system for detecting and matching anatomical structures in an image to one or more anatomical structures in a training set of images comprising the steps of:

means for receiving a candidate image;

means for extracting feature values from the candidate image;

means for applying a classification function to detect an anatomical structure;

means for identifying one or more counterpart images in the training set of images by matching the extracted feature values for the candidate image to feature values of the counterpart images in the training set; and

means for using a shape of an anatomical structure in a matching counterpart image from the training set to determine a shape of the anatomical structure in the candidate image.

24. The system of claim 23 wherein the means for identifying counterpart images further comprises:

means for generating a feature vector for the candidate anatomical structure; and

means for comparing the candidate feature vector to a feature vector associated with a counterpart image.

25. The system of claim 24 wherein the feature vector comprises a plurality of weak classifier outputs h_i'"'"'s and associated weights a_i'"'"'s.

26. The system of claim 23 wherein said means for applying a classification function further comprises:

means for scanning a window through the candidate image.

27. The system of claim 23 wherein the classification function uses boosting.

28. The system of claim 23 wherein the classification function uses AdaBoosting.

29. The system of claim 23 wherein the classification function uses RankBoosting.

30. The system of claim 23 wherein the classification function uses a support vector machine.

31. The system of claim 23 wherein the image is an ultrasound image.

32. The system of claim 31 wherein the anatomical structure is a left ventricle.

33. The system of claim 23 wherein the anatomical structure is a face.