Method and apparatus for tracking motion in MR images

• US 6,292,683 B1
• Filed: 05/18/1999
• Issued: 09/18/2001
• Est. Priority Date: 05/18/1999
• Status: Expired due to Term

• Current Assignee: General Electric Company, Sunnybrook and Women's College Health Sciences Centre
• Original Assignee: General Electric Company
• Inventors: Wright, Graham, Gupta, Sandeep Narendra, Sussman, Marshall
• Primary Examiner(s): Casler, Brian L.
• Application Number:
  US09/313,910
• Time in Patent Office: 854 Days
• Field of Search: 324/307, 324/309, 324/306, 600/410, 600/416, 600/413
• US Class Current: 600/410
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: G06T 3/147 using affine transformations

Abstract

The present invention relates to magnetic resonance imaging (MRI) and includes a method and apparatus to track motion of anatomy or medical instruments, for example, between MR images. The invention includes acquiring a time series of MR images of a region of interest, where the region of interest contains the anatomy or structure that is prone to movement, and the MR images contain signal intensity variations. The invention includes identifying a local reference region in the region of interest of a reference image and acquired from the time series. The local reference region of the reference image is compared to that of the other MR images and a translational displacement is determined between the local reference region of the reference image and of another MR image. The translational displacement has signal intensity invariance and can accurately track anatomy motion or the movement of a medical instrument during an invasive procedure. The translational displacement can be used to align the images for automatic registration, such as in myocardial perfusion imaging, MR angiography, fMRI, or in any other procedure in which motion tracking is advantageous. Two implementations of the invention are disclosed, one in which a correlation coefficient is calculated and used to determine the translational displacement, and one in which the images are converted to a binary image by thresholding and after computation of a filtered cross-correlation, a signal peak is located and plotted as the translational displacement.

Claims

1. A method of tracking motion in a series of MR images comprising the steps of:

acquiring a time series of MR images of a region of interest of a patient, where the region of interest contains structure that is prone to movement during imaging and where the series of MR images contain signal intensity variations;

identifying a local reference region in the region of interest of a reference image from the series of MR images;

comparing the local reference region of the reference image to that of another MR image in the series of MR images; and

determining a translational displacement between the local reference region of the reference image and another MR image, the translational displacement having a signal intensity invariance to accurately track structure movement in the region of interest.

2. The method of claim 1 further comprising the step of repeating the comparing and determining steps for each MR image acquisition until a steady state estimate of translational displacement is determined.

3. The method of claim 1 wherein the structure is anatomy and the method further comprises the step of applying the translational displacement to align the MR images of the anatomy in the series of MR images for automatic registration, thereby compensating for anatomy movement between MR images.

4. The method of claim 1 wherein the region of interest contains a patient'"'"'s cardiac system and wherein the method includes the step of correcting MR image displacement due to respiratory movement to accomplish automatic image registration for cardiac MRI perfusion analysis.

5. The method of claim 1 wherein the step of determining a translational displacement includes converting each MR image to a binary image.

6. The method of claim 5 further comprising the steps of creating a cross-correlation image from the binary images of a previous MR image and a subsequent MR image and locating a peak of the cross-correlation image.

7. The method of claim 6 further comprising the step of computing a shift factor in an x-direction and in a y-direction from the cross-correlation image peak.

8. The method of claim 7 further comprising the step of shifting the subsequent MR image by the shift factor to align the subsequent MR image with the previous MR images.

9. The method of claim 8 further comprising the step of repeating the steps of creating a cross-correlation image, locating a peak of the cross-correlation image, computing a shift factor, and shifting the subsequent MR image for each MR image in the series of MR images such that all the MR images in the series of MR images are automatically registered.

10. The method of claim 8 further comprising the step of repeating the steps of creating a cross-correlation image, locating a peak of the cross-correlation image, and computing a shift factor until the shift factor is equal to zero as between the previous MR image and the subsequent MR image.

11. The method of claim 5 wherein the step of converting each MR image to a binary image is further defined as determining an intensity threshold for each MR image.

12. The method of claim 11 wherein the step of determining an intensity threshold is further defined as obtaining an average pixel intensity for each image, and assigning a binary 1 to all pixel values above the average pixel intensity, and assigning a binary 0 to all pixel values below the average pixel intensity.

13. The method of claim 1 further comprising the step of calculating a correlation coefficient between each MR image and the reference image at a selected location and determining the translational displacement based on the correlation coefficient.

14. The method of claim 13 wherein the step of calculating a correlation coefficient is based on a least squares distance function.

15. The method of claim 13 further comprising the step of selecting a location having a largest correlation coefficient value across regions in an MR image.

16. The method of claim 15 further comprising the step of restricting the selection of a location based on a tolerance defined by the location and size of a previously selected larger pattern, at a location where the correlation coefficient is maximum.

17. The method of claim 13 wherein the correlation coefficient provides an indication of similarity of a selected region of the reference image with that of another MR image.

18. The method of claim 13 further comprising the step of aligning a selected image with the reference image based on the correlation coefficient calculated.

19. The method of claim 13 further comprising the steps of determining an accuracy of the translational displacement determined and rejecting the translational displacement on the accuracy determined.

20. The method of claim 19 wherein the step of determining an accuracy of the translational displacement includes analyzing the correlation coefficient against a probability of obtaining a value of the correlation coefficient that is greater than a threshold value r₀ and the step of rejecting the translational displacement is further defined as rejecting the translational displacement where the value of the correlation coefficient is less than the threshold value r₀.

21. The method of claim 19 wherein the step of determining an accuracy of the translational displacement includes a time series analysis of a movement parameter and the step of rejecting the translational displacement is based on whether the movement parameter exceeds a given tolerance.

22. The method of claim 1 wherein the region of interest is a patient'"'"'s coronary artery and wherein the method includes the step of correcting MR image displacement to accomplish automatic image registration for MR angiography.

23. The method of claim 1 wherein the structure is a medical instrument and further comprises the step of tracking movement of the medical instrument during an invasive procedure.

24. An MRI apparatus to track structure movement in a series of MR images comprising:

a magnetic resonance imaging (MRI) system having a plurality of gradient coils positioned about a bore of a magnet to impress a polarizing magnetic field and an RF transceiver system having an RF modulator controlled by a pulse control module to transmit RF signals to an RF coil assembly to acquire MR images; and

a computer programmed to;

periodically activate the MRI system and acquire a time series of MR images having signal intensity variations, each MR image acquiring a region of interest of a patient containing structure that is prone to movement;

identify a local reference region in the region of interest of a reference image from the series of MR images; and

determine a translational displacement of the local reference region in the reference image as compared to another MR image to track structure movement in the region of interest occurring between MR image acquisitions, the translational displacement having signal intensity invariance such that the translational displacement is determinable and capable of providing tracking data of the structure movement occurring between MR image acquisitions regardless of signal intensity variations within a given MR image and between MR images.

25. The MRI apparatus of claim 24 wherein the structure is patient anatomy and the computer is further programmed to apply the translational displacement to align the MR images in the series of MR images to compensate for anatomy movement and accomplish automatic registration.

26. The MRI apparatus of claim 24 wherein the computer is further programmed to determine the translational displacement by converting a displaced MR image to a black and white binary image.

27. The MRI apparatus of claim 26 wherein the computer is further programmed to create a cross-correlation image from the black and white binary image and the reference image, and locate a peak of the cross-correlation image.

28. The MRI apparatus of claim 27 wherein the computer is further programmed to:

compute a shift factor in an x-direction and in a y-direction from the cross-correlation image peak; and

shift the displaced MR image by the shift factor to align the displaced MR image with the reference images.

29. The MRI apparatus of claim 26 wherein the computer is further programmed to:

determine an intensity threshold for each MR image by calculating an average pixel intensity for each image;

assign a binary 1 to all pixel values above the average pixel intensity; and

assign a binary 0 to all pixel values below the average pixel intensity.

30. The MRI apparatus of claim 24 wherein the computer is further programmed to continuously repeat the translational displacement determination for each MR image acquisition until a steady state estimate of translational displacement is determined.

31. The MRI apparatus of claim 24 wherein the computer is further programmed to calculate a correlation coefficient between each image from the series of MR images and the reference image at a selected location and determine the translational displacement based on the correlation coefficient.

32. The MRI apparatus of claim 31 wherein the computer is further programmed to separately calculate a numerator and a denominator of the correlation coefficient to efficiently calculate the translational displacement.

33. The MRI apparatus of claim 31 wherein the computer is further programmed to determine the translational displacement by calculating a least squared distance between the reference image and another MR image from the series of MR images.

34. The MRI apparatus of claim 31 wherein the computer is further programmed to select a location of a pattern match based on identifying a location having a maximum correlation coefficient value across an MR image.

35. The MRI apparatus of claim 31 wherein the computer is further programmed to provide an indication of location similarity between MR images based on the value of the correlation coefficient.

36. The MRI apparatus of claim 31 wherein the computer is further programmed to align a selected image with the reference image based on the correlation coefficient calculated.

37. The MRI apparatus of claim 31 wherein the computer is further programmed to reject any translational displacement determination based upon a probability of obtaining a correlation coefficient value that is less than a predetermined threshold value.

38. The MRI apparatus of claim 31 wherein the computer is further programmed to reject any translational displacement determination based upon a comparison of a translational displacement parameter to a given tolerance, the translational displacement parameter being one of:

a total maximum displacement parameter, a velocity parameter, and an acceleration parameter, each a parameter of the structure movement in the region of interest.

39. A computer system for use with an MRI apparatus comprising a computer programmed from a computer readable storage medium having thereon a computer program programmed to:

acquire a time series of MR images of a region of interest of a patient, where the region of interest contains structure that is prone to movement and where the MR images are capable of having signal intensity variations;

select a reference image from the series of MR images;

identify a local reference region in the region of interest for the reference image from the series of MR images;

compare the local reference region in the reference image to other MR images in the series of MR images; and

determine a translational displacement between a selected MR image and the reference image that is independent and unaffected by the signal intensity variations such that the translational displacement can be used to track movement of the structure in the region of interest between MR images.

40. The computerized system of claim 39 wherein the computer is further programmed to repeat the translational displacement determination between a next selected MR image and a previous MR image for each MR image acquisition until a steady state estimate of translational displacement is determined and each MR image in the series of MR images have the region of interest in alignment.

41. The computerized system of claim 39 wherein the structure is anatomy and the computer is further programmed to apply the measure of translational displacement to shift the selected MR image into alignment with the reference image and align the MR images in the series of MR images for automatic registration and thereby correct for anatomy movement in the region of interest.

42. The computerized system of claim 39 wherein the region of interest includes at least a portion of a heart and the computer is further programmed to correct MR image displacement due to respiratory movement of the heart to accomplish automatic image registration for cardiac MRI perfusion analysis.

43. The computerized system of claim 39 wherein the computer is further programmed to convert each MR image to a binary image to determine a translational displacement.

44. The computerized system of claim 43 wherein the computer is further programmed to create a cross-correlation image from the binary images of a previous MR image and from a subsequent MR image, and to locate a peak of the cross-correlation image.

45. The computerized system of claim 44 wherein the computer is further programmed to compute a shift factor in an x-direction and in a y-direction from the cross-correlation image peak.

46. The computerized system of claim 45 wherein the computer is further programmed to shift the subsequent MR image by the shift factor to align the subsequent MR image with the previous MR image.

47. The computerized system of claim 46 wherein the computer is further programmed to continuously repeat the creation of a cross-correlation image, locating a peak of the cross-correlation image, computing a shift factor, and shifting the subsequent image for each MR image in the series of MR images such that all the MR images in the series of MR images are automatically registered.

48. The computerized system of claim 46 wherein the computer is further programmed to continuously repeat the steps of creating a cross-correlation image, locating a peak of the cross-correlation image, and computing a shift factor between a given MR image and a given reference image until the shift factor is equal to zero.

49. The computerized system of claim 43 wherein the computer is further programmed to determine an intensity threshold for each MR image by converting each MR image to a binary image.

50. The computerized system of claim 49 wherein the computer is further programmed to obtain an average pixel intensity for each image, assign a binary 1 to all pixel values at and above the average pixel interest, and to assign a binary 0 to all pixel values below the average pixel intensity.

51. The computerized system of claim 39 wherein the computer is further programmed to calculate a correlation coefficient between each MR image and the reference image at a selected location and determine the translational displacement based on the correlation coefficient.

52. The computerized system of claim 51 wherein the computer is further programmed to calculate a correlation coefficient based on a least squares distance function.

53. The computerized system of claim 51 wherein the computer is further programmed to select a location for a pattern match based on identifying the correlation coefficient having a largest correlation coefficient value across an MR image.

54. The computerized system of claim 53 wherein the computer is further programmed to restrict a location search based on a tolerance defined by the location and size of a previously selected larger pattern, at a location where the correlation coefficient is maximum.

55. The computerized system of claim 51 wherein the computer is further programmed to provide an indication of similarity of location of the reference image with that of another MR image.

56. The computerized system of claim 51 wherein the computer is further programmed to align a selected image with the reference image based on the correlation coefficient calculated.

57. The computerized system of claim 51 wherein the computer is further programmed to determine an accuracy of the translational displacement determined, and accept the translational displacement based on the accuracy determined.

58. The computerized system of claim 57 wherein the computer is further programmed to analyze the correlation coefficient against a probability of obtaining a value of the correlation coefficient that is greater than a threshold value r₀ and reject a value of the translational displacement that is less than the threshold value r₀.

59. The computerized system of claim 51 wherein the computer is further programmed to determine an accuracy of the translational displacement that includes a time series analysis of an anatomy movement parameter and to reject any translational displacement that exceeds a given tolerance of the anatomy movement parameter.

60. The computerized system of claim 39 wherein the computer is further programmed to track movement of a medical instrument during an invasive procedure.