System and method for medical image correction

• US 9,953,397 B2
• Filed: 09/05/2014
• Issued: 04/24/2018
• Est. Priority Date: 09/05/2014
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Tan, Ek Tsoon, Liu, Xiaofeng, Xu, Dan
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): TORRES, JOSE
• Application Number:
  US14/478,597
• Publication Number:
  US 20160071269A1
• Time in Patent Office: 1,327 Days
• Field of Search: None
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 3/00 Geometric image transformations in the plane of the image
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: G06T 7/32 using correlation-based methods

Abstract

A method implemented using at least one processor includes receiving a target image and a reference image. The target image is a distorted magnetic resonance image and the reference image is an undistorted magnetic resonance image. The method further includes selecting an image registration method for registering the target image to the reference image, wherein the image registration method uses an image transformation. The method further includes performing image registration of the target image with the reference image, wherein the image registration provides a plurality of optimized parameters of the image transformation. The method also includes generating a corrected image based on the target image and the plurality of optimized parameters of the image transformation.

Claims

1. A method, comprising:

receiving a target image, wherein the target image is a distorted magnetic resonance image comprising distortions in y-coordinates of pixels of the target image based on application of phase encoding gradient pulse along a y-axis and frequency encoding gradient pulse along an x-axis in an interleaved fashion;

receiving a reference image, wherein the reference image is an undistorted magnetic resonance image;

selecting an image registration method for registering the target image to the reference image, wherein the image registration method uses an image transformation to reduce distortions, wherein the image transformation consists of modifying the y-coordinates of the pixels based on corresponding pixels of the reference image using a polynomial function;

performing image registration of the target image with the reference image, wherein the image registration provides a plurality of optimized parameters of the image transformation; and

generating a corrected image based on the target image and the plurality of optimized parameters of the image transformation.

2. The method of claim 1, wherein receiving the target image comprises acquiring a magnetic resonance image by a fast sampling technique.

3. The method of claim 1, wherein receiving the target image comprises acquiring a magnetic resonance image by an echo planar imaging technique.

4. The method of claim 1, wherein the distortions in the y-coordinates of the pixels of the target image comprises image distortion due to magnetic field variations.

5. The method of claim 1, wherein performing the image registration comprises transforming the target image based on a non-rigid transformation.

6. The method of claim 5, wherein the non-rigid transformation is based on a third order non-linear polynomial function.

7. The method of claim 1, wherein performing the image registration comprises determining a similarity value based on a similarity metric.

8. The method of claim 7, wherein determining the similarity value comprises computation of a normalized mutual information.

9. The method of claim 1, wherein performing the image registration comprises determining the plurality of optimized parameters of the image transformation based on an optimization technique.

10. The method of claim 9, wherein the optimization technique comprises a gradient based method.

11. A system, comprising:

at least one processor and a memory communicatively coupled to a communications bus;

an image acquisition module for receiving a target image and a reference image, wherein the target image is a distorted magnetic resonance image and the reference image is an undistorted magnetic resonance image, and wherein the target image comprises distortions in y-coordinates of pixels based on application of phase encoding gradient pulse along a y-axis and frequency encoding gradient pulse along an x-axis in an interleaved fashion;

an image registration module communicatively coupled to the image acquisition module and configured to;

select an image registration method for registering the target image to the reference image, wherein the image registration method uses an image transformation to reduce distortions, wherein the image transformation consists of modifying the y-coordinates of the pixels of the target image based on corresponding pixels of the reference image using a polynomial function;

perform image registration of the target image with the reference image, wherein the image registration provides a plurality of optimized parameters of the image transformation; and

an image correction module communicatively coupled to the image registration module and configured to generate a corrected image based on the target image and the plurality of optimized parameters of the image transformation,

wherein the image acquisition module, the image registration module, and the image correction module are coupled to the communications bus, and wherein at least one of the image acquisition module, the image registration module, and the image correction module is stored in the memory and executable by the at least one processor.

12. The system of claim 11, wherein the image acquisition module is configured to acquire the target image using a fast sampling technique.

13. The system of claim 11, wherein the image acquisition module is configured to acquire the target image using an echo planar imaging technique.

14. The system of claim 11, wherein the image acquisition module introduces image distortion due to magnetic field variations.

15. The system of claim 11, wherein the image registration module is configured to perform a non-rigid transformation of the target image.

16. The system of claim 15, wherein the image registration module is configured to transform the target image based on a third order non-linear polynomial function.

17. The system of claim 11, wherein the image registration module is further configured to determine a similarity value based on a similarity metric.

18. The system of claim 17, wherein the image registration module is further configured to determine a normalized mutual information.

19. The system of claim 11, wherein the image registration module is further configured to determine the plurality of optimized parameters of the image transformation based on an optimization technique.

20. The system of claim 19, wherein the image registration module is further configured to perform a gradient based optimization technique.

21. A non-transitory computer readable medium encoded with a program to enable at least one processor to:

receive a target image, wherein the target image is a distorted magnetic resonance image comprising distortions in y-coordinate of pixels of the target image based on application of phase encoding gradient pulse along a y-axis and frequency encoding gradient pulse along x-axis in an interleaved fashion;

receive a reference image, wherein the reference image is an undistorted magnetic resonance image;

select an image registration method for registering the target image to the reference image, wherein the image registration method uses an image transformation to reduce distortions, wherein the image transformation consists of modifying the y-coordinates of the pixels based on corresponding pixels of the reference image using a polynomial function;

perform image registration of the target image with the reference image, wherein the image registration provides a plurality of optimized parameters of the image transformation; and

generate a corrected image based on the target image and the plurality of the optimized parameters of the image transformation.