System and method for gradient assisted non-connected automatic Region (GANAR) analysis

• US 20150055849A1
• Filed: 08/21/2013
• Published: 02/26/2015
• Est. Priority Date: 08/21/2013
• Status: Active Grant

• Current Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Original Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Inventors: Jeraj, Robert, Galavis, Paulina E., Paliwal, Bhudatt, Holden, James
• Granted Patent: US 9,355,447 B2
• Time in Patent Office: Days
• US Class Current: 382/132
• CPC Class Codes: A61N 5/103 Treatment planning systems
• CPC Class Codes: G06T 2207/10104 Positron emission tomography [PET]
• CPC Class Codes: G06T 2207/10108 Single photon emission computed tomography [SPECT]
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/155 involving morphological operators
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/41 based on statistical description of texture

Abstract

A system and method for analyzing medical images of a subject includes acquiring the medical images of the subject and texture images. A computer system determines, using the medical images and the texture feature images, a plurality of segmentation surfaces by iteratively adjusting a relationship between a region growing algorithm that selects a region of interest (ROI) to determine a given segmentation surface and cost function for evaluating the given segmentation surface. The computer system generates a report using the plurality of segmentation surfaces indicating at least boundaries between anatomical structures with functional differences in the medical images.

Claims

1. A method for programming operation of a radiation therapy system to deliver radiation therapy to a subject, the method comprising:

Acquiring medical images of the subject including functional information and anatomical information about the subject;

Acquiring texture feature images;

using the medical images and the texture feature images, determining a plurality of segmentation surfaces by minimizing a relationship between a region growing algorithm that selects a region of interest (ROI) to determine a given segmentation surface and cost function for evaluating the given segmentation surface;

Synthesizing the plurality of segmentation surfaces into a segmentation report; and

using information from the segmentation report to program operation of the radiation therapy system for a delivery of radiation therapy to the subject.

2. The method of claim 1 wherein the relationship between the region growing algorithm and the cost function is defined by a single variable.

3. The method of claim 2 wherein the single variable is an inhomogeneity parameter.

4. The method of claim 1 wherein the relationship between the region growing algorithm and the cost function is defined by an inhomogeneity parameter that is used by the region growing algorithm to determine a largest segmentation surface that yield a degree of inhomogeneity less than a maximum acceptable measure of inhomogeneity and is used by the cost function to evaluate the largest segmentation surface to control inter-voxel and intra-voxel variances relative to the largest segmentation surface.

5. The method of claim 4 wherein the inhomogeneity parameter is disregarded when synthesizing the plurality of segmentation surfaces into a segmentation report.

6. The method of claim 1 wherein the segmentation report forms a portion of a radiation therapy plan.

7. The method of claim 1 wherein synthesizing the plurality of segmentation surfaces into a segmentation report includes calculating a probabilistic estimate of a probabilistic segmentation surface from the plurality of segmentation surfaces.

8. The method of claim 1 wherein the synthesizing the plurality of segmentation surfaces into a segmentation report includes performing an expectation maximization algorithm.

9. The method of claim 8 wherein the expectation maximization algorithm includes a simultaneous truth and performance level estimation (STAPLE) algorithm.

10. The method of claim 1 wherein the texture feature images are entire texture feature images.

11. A system for processing medical images of a subject, the system comprising:

a communications connection configured to acquire medical images of the subject including functional information and anatomical information about the subject;

a non-transitory storage medium having stored thereon texture feature images;

a computer system configured to;

communicate with the communications connection to receive the medical images;

communicate with the non-transitory storage medium to access the texture images;

determine, using the medical images and the texture feature images, a plurality of segmentation surfaces by iteratively adjusting a relationship between a region growing algorithm that selects a region of interest (ROI) to determine a given segmentation surface and cost function for evaluating the given segmentation surface; and

generate a report using the plurality of segmentation surfaces indicating at least boundaries between anatomical structures with functional differences in the medical images.

12. The system of claim 11 wherein the computer system is further configured to synthesize the plurality of segmentation surfaces into a segmentation report.

13. The system of claim 12 wherein the computer system is further configured to calculate a probabilistic estimate of a probabilistic segmentation surface from the plurality of segmentation surfaces when synthesizing the plurality of segmentation surfaces into a segmentation report.

14. The system of claim 12 wherein the computer system is further configured to use information from the segmentation report to program operation of a radiation therapy system for a delivery of radiation therapy to the subject.

15. The system of claim 12 wherein the computer system is further configured to perform an expectation maximization algorithm when the synthesizing the plurality of segmentation surfaces into a segmentation report.

16. The system of claim 15 wherein the expectation maximization algorithm includes a simultaneous truth and performance level estimation (STAPLE) algorithm.

17. The system of claim 11 wherein the relationship between the region growing algorithm and the cost function is defined by a single variable.

18. The system of claim 17 wherein the single variable is an inhomogeneity parameter that is used by the region growing algorithm to determine a largest segmentation surface that yield a degree of inhomogeneity less than a maximum acceptable measure of inhomogeneity and is used by the cost function to evaluate the largest segmentation surface to control inter-voxel and intra-voxel variances relative to the largest segmentation surface.

19. The system of claim 17 wherein the single variable is disregarded when generating the report.

20. The system of claim 17 wherein the computer system is configured to use the single variable as a stopping criteria when determining the plurality of segmentation surfaces.