AUTOMATIC 3-D SEGMENTATION OF THE SHORT-AXIS LATE-ENHANCEMENT CARDIAC MRI

• US 20100303328A1
• Filed: 01/26/2009
• Published: 12/02/2010
• Est. Priority Date: 01/31/2008
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Ciofolo, Cybele, Mory, Benoit, Fradkin, Maxim
• Granted Patent: US 8,509,506 B2
• Time in Patent Office: Days
• US Class Current: 382/131
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• 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

Abstract

The invention relates to a system (200) for delineating an anatomical structure in an image computed from a slice of image data, the system (200) comprising a partition unit (210) for partitioning the image into a plurality of image portions, each image portion depicting a portion of the anatomical structure, and an adaptation unit (220) for adapting a template to the image based on a criterion function, the criterion function being a function of template parameters and of image values and their relative positions in the image, and on a criterion to be satisfied by a computed value of the criterion function, wherein the criterion function is defined on the basis of the plurality of image portions. Having a criterion function defined on the basis of the plurality of image portions allows computing optimal contributions to the criterion function value within each portion of the image. This is proven to be particularly useful for extracting myocardial contours in a slice of the LEMR short-axis volume image data in the presence of scar tissues. The invention allows optimizing the treatment of portions of the image comprising scar tissue in the myocardium.

Claims

1. A system (200) for delineating an anatomical structure in an image computed from a slice of image data, the system (200) comprising:

a partition unit (210) for partitioning the image into a plurality of image portions, each image portion depicting a portion of the anatomical structure; and

an adaptation unit (220) for adapting a template to the image, based on a criterion function, the criterion function being a function of template parameters and of image values and their relative positions in the image, and on a criterion to be satisfied by a computed value of the criterion function,

wherein the criterion function is defined on the basis of the plurality of image portions.

2. A system (200) as claimed in claim 1, wherein the adaptation unit (220) is further arranged to detect an abnormal tissue of the anatomical structure in an image portion of the plurality of image portions and the criterion function comprises a term contributing to the criterion function when said abnormal tissue is detected in said image portion.

3. A system (200) as claimed in claim 1, wherein adapting the template is based on searching for values of the template parameters, for which the criterion function satisfies the criterion condition, using a greedy search algorithm.

4. A system (200) as claimed in claim 1, wherein the template is a closed ribbon defined by an external and an internal closed contour.

5. A system as claimed in claim 1, wherein the partition is a pie-like partition.

6. A system (200) as claimed in claim 1, further comprising a registration unit (230) for registering a surface model with a plurality of templates, wherein each template is adapted to an image computed from a slice of image data and wherein at least one template is adapted by the adaptation unit (220) of the system (200).

7. A system (200) as claimed in claim 6, wherein the registration unit (230) comprises:

an affine transformation unit (232) for affinely registering the surface model with the plurality of templates, thereby creating an affinely registered surface model;

a local deformation unit (234) for locally non-affinely registering the affinely registered surface model with the plurality of templates, thereby creating a locally registered surface model; and

a refinement unit (236) for adapting the locally registered surface model to the image data.

9. An image acquisition apparatus (800) comprising a system (200) as claimed in claim 1.

10. A workstation (900) comprising a system (200) as claimed in claim 1.

8. A method (700) of delineating an anatomical structure in an image computed from a slice of image data, the method (700) comprising:

a partition step (710) for partitioning the image into a plurality of image portions, each image portion depicting a portion of the anatomical structure; and

an adaptation step (720) for adapting a template to the image, based on a criterion function, the criterion function being a function of template parameters and of image values and their relative positions in the image, and on a criterion to be satisfied by a computed value of the criterion function,

wherein the criterion function is defined on the basis of the plurality of image portions.

11. A computer program product to be loaded by a computer arrangement, comprising instructions for delineating an anatomical structure in an image computed from a slice of image data, the computer arrangement comprising a processing unit and a memory, the computer program product, after being loaded, providing said processing unit with the capability to carry out the tasks of:

partitioning the image into a plurality of image portions, each image portion depicting a portion of the anatomical structure; and

adapting a template to the image, based on a criterion function, the criterion function being a function of template parameters and of image values and their relative positions in the image, and on a criterion to be satisfied by a computed value of the criterion function,

wherein the criterion function is defined on the basis of the plurality of image portions.