Method and apparatus for brain perfusion magnetic resonance images

• US 8,427,151 B2
• Filed: 01/14/2011
• Issued: 04/23/2013
• Est. Priority Date: 01/14/2011
• Status: Expired due to Fees

• Current Assignee: National Yang-Ming University
• Original Assignee: National Yang-Ming University
• Inventors: Zheng, Wen-Yan, Teng, Mu-Huo, Kao, Yi-Hsuan, Chen, Yu-Fen
• Primary Examiner(s): SHRIVASTAV, BRIJ B
• Application Number:
  US13/007,357
• Publication Number:
  US 20120184843A1
• Time in Patent Office: 830 Days
• Field of Search: 324300-322, 600407-445
• US Class Current: 324/307
• CPC Class Codes: A61B 5/0042 for the brain
• CPC Class Codes: A61B 5/0263 using NMR
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: A61B 5/4064 Evaluating the brain A61B5/031, A61B5/369, A61B5/14553 take precedence
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G06T 2207/10096 Dynamic contrast-enhanced magnetic resonance imaging [DCE-MRI]
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/20224 Image subtraction
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 7/0016 involving temporal comparison
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06T 7/174 involving the use of two or more images

Abstract

The present invention discloses a method and apparatus for brain perfusion magnetic resonance imaging (MRI) technique with the removal of cerebrospinal fluid (CSF) pixels. This invention utilizes a CSF/brain-contrast-enhanced image, wherein the CSF/brain-contrast-enhanced image is defined as the signal difference between CSF and brain matter divided by a standard deviation of air background random noise is larger than 3, acquired from the subject'"'"'s brain, and applies a segmentation technique to remove the CSF pixels. After removing the CSF pixels on parametric images, the extent of brain tissue with delayed perfusion can be better identified. By using a good region of interest enclosing the correct delayed-perfusion region, the measurement on the tissue volume and perfusion parameters would be more accurate than the area contaminated by CSF pixels.

Claims

1. A method for removing cerebrospinal fluid (CSF) pixels on a brain perfusion parametric image of a subject, with a magnetic resonance imaging system, the subject being administered with a contrast agent, comprising the steps of:

a) acquiring a series of dynamic magnetic resonance images from the subject'"'"'s brain wherein the dynamic magnetic resonance images comprises at least a signal of CSF and at least a signal of brain matter;

b) calculating a brain perfusion parametric image from the dynamic images;

c) acquiring a CSF/brain-contrast-enhanced image, wherein the CSF/brain-contrast-enhanced image is defined as the signal difference between CSF and brain matter divided by a standard deviation of air background random noise is larger than 3, from the subject'"'"'s brain;

d) applying a segmentation method to the CSF/brain-contrast-enhanced image with or without the use of other images of the subject'"'"'s brain for identifying a plurality of CSF pixels containing the signal of CSF; and

e) removing the CSF pixels identified in the step d), when displaying the brain perfusion parametric image calculated in the step b).

2. The method as claimed in claim 1, wherein in the step c), acquiring the CSF/brain-contrast-enhanced images can be performed either separately or continuously with the dynamic images.

3. The method as claimed in claim 1, wherein acquiring the CSF/brain-contrast-enhanced image can be performed using a pulse sequence with an effective repetition time (TR) longer than 2 seconds, in which CSF is bright on the images.

4. The method as claimed in claim 2, wherein acquiring the CSF/brain-contrast-enhanced image can be performed using a pulse sequence with an effective repetition time (TR) longer than 2 seconds, in which CSF is bright on the images.

5. The method as claimed in claim 1, wherein acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signal of CSF, in which CSF is dark on the image.

6. The method as claimed in claim 2, wherein acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signal of CSF, in which CSF is dark on the image.

7. The method as claimed in claim 1, wherein acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signals of brain matter, in which brain matter is dark on the image.

8. The method as claimed in claim 2, wherein acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signals of brain matter, in which brain matter is dark on the image.

9. The method as claimed in claim 1, wherein in the step d), a thresholding technique is used in the segmentation method.

10. The method as claimed in claim 1, wherein in the step d), a clustering technique is used in the segmentation method.

11. The method as claimed in claim 1, wherein in the step e), a factor analysis technique is used in the segmentation method.

12. The method as claimed in claim 11, wherein a thresholding technique is applied to an output image of the factor analysis technique.

13. The method as claimed in claim 11, wherein a clustering technique is applied to an output images of the factor analysis technique.

14. The method as claimed in claim 1, wherein the step d) further comprises the step of compensating the CSF/brain-contrast-enhanced image to be an inhomogeneity compensated image for the inhomogeneity of excitation radiofrequency field.

15. The method as claimed in claim 14, wherein the step of compensating is achieved by using a reference image, wherein the reference image is calculated by averaging the Pth to the Qth dynamic images, and P and Q are both integers.

16. The method as claimed in claim 14, wherein a thresholding technique is applied to the inhomogeneity compensated image.

17. The method as claimed in claim 14, wherein a clustering technique is applied to the inhomogeneity compensated image.

18. The method as claimed in claim 1, wherein in the step b), the calculated brain perfusion parameter image is at least one of cerebral blood volume (CBV), cerebral blood flow (CBF), relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), time-to-peak (TTP), and mean transit time (MTT) images.

19. The method as claimed in claims 9, wherein the thresholding technique is one of global thresholding technique, or local thresholding technique.

20. The method as claimed in claims 12, wherein the thresholding technique is one of global thresholding technique, or local thresholding technique.

21. The method as claimed in claims 16, wherein the thresholding technique is one of global thresholding technique, or local thresholding technique.

22. The method as claimed in claim 10, wherein the clustering technique is one of k-mean clustering technique, or fuzzy c-means clustering technique, or Bayes classifier technique, or Markov random field technique.

23. The method as claimed in claim 13, wherein the clustering technique is one of k-mean clustering technique, or fuzzy c-means clustering technique, or Bayes classifier technique, or Markov random field technique.

24. The method as claimed in claim 17, wherein the clustering technique is one of k-mean clustering technique, or fuzzy c-means clustering technique, or Bayes classifier technique, or Markov random field technique.

25. The method as claimed in claim 11, wherein the factor analysis technique is one of principle component analysis or independent component analysis.

26. The method as claimed in claim 15, wherein P is equal to six and Q is equal to ten.

27. An apparatus for removing cerebrospinal fluid (CSF) pixels on a brain perfusion parametric image of a subject, with a magnetic resonance imaging system, the subject being administered with a contrast agent, comprising:

a) first acquiring mean for acquiring a series of dynamic magnetic resonance images from the subject'"'"'s brain, wherein the dynamic magnetic resonance images comprises at least a signal of CSF and at least a signal of brain matter;

b) calculating mean for calculating a brain perfusion parametric image from the dynamic images;

c) second acquiring mean for acquiring a CSF/brain-contrast-enhanced image, wherein the CSF/brain-contrast-enhanced image is defined as the signal difference between CSF and brain matter divided by a standard deviation of air background random noise is larger than 3, from the subject'"'"'s brain;

d) applying mean for applying a segmentation method to the CSF/brain-contrast-enhanced image with or without the use of other images of the subject'"'"'s brain for identifying a plurality of CSF pixels containing the signal of CSF; and

e) removing mean for removing the CSF pixels identified from the applying mean, when displaying the brain perfusion parametric image calculated in the calculating mean.

28. The apparatus as claimed in claim 27, wherein mean for acquiring the CSF/brain-contrast-enhanced images can be performed either separately or continuously with the dynamic images.

29. The apparatus as claimed in claim 27, wherein mean for acquiring the CSF/brain-contrast-enhanced image can be performed using a pulse sequence with an effective repetition time (TR) longer than 2 seconds, in which CSF is bright on the images.

30. The apparatus as claimed in claim 28, wherein mean for acquiring the CSF/brain-contrast-enhanced image can be performed using a pulse sequence with an effective repetition time (TR) longer than 2 seconds, in which CSF is bright on the images.

31. The apparatus as claimed in claim 27, wherein mean for acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signal of CSF, in which CSF is dark on the image.

32. The apparatus as claimed in claim 28, wherein mean for acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signal of CSF, in which CSF is dark on the image.

33. The apparatus as claimed in claim 27, wherein mean for acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signals of brain matter, in which brain matter is dark on the image.

34. The apparatus as claimed in claim 28, wherein mean for acquiring the CSF/brain-contrast-enhanced image can be performed using an inversion-recovery pulse sequence with an inversion time used to reduce the signals of brain matter, in which brain matter is dark on the image.

35. The apparatus as claimed in claim 27, wherein a thresholding technique is used in the segmentation method.

36. The apparatus as claimed in claim 27, wherein a clustering technique is used in the segmentation method.

37. The apparatus as claimed in claim 27, wherein a factor analysis technique is used in the segmentation method.

38. The apparatus as claimed in claim 37, wherein a thresholding technique is applied to an output image of the factor analysis technique.

39. The apparatus as claimed in claim 37, wherein a clustering technique is applied to an output images of the factor analysis technique.

40. The apparatus as claimed in claim 27, wherein mean for applying a segmentation method compensates the CSF/brain-contrast-enhanced image to be an inhomogeneity compensated image for the inhomogeneity of excitation radiofrequency field.

41. The apparatus as claimed in claim 40, wherein mean for applying a segmentation method compensates the CSF/brain-contrast-enhanced image by using a reference image, wherein the reference image is calculated by averaging the Pth to the Qth dynamic images, and P and Q are both integers.

42. The apparatus as claimed in claim 40, wherein a thresholding technique is applied to the inhomogeneity compensated image.

43. The apparatus as claimed in claim 40, wherein a clustering technique is applied to the inhomogeneity compensated image.

44. The apparatus as claimed in claim 27, wherein in the step b), the calculated brain perfusion parameter image is at least one of cerebral blood volume (CBV), cerebral blood flow (CBF), relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), time-to-peak (TTP), and mean transit time (MTT) images.

45. The apparatus as claimed in claims 35, wherein the thresholding technique is one of global thresholding technique, or local thresholding technique.

46. The apparatus as claimed in claims 38, wherein the thresholding technique is one of global thresholding technique, or local thresholding technique.

47. The apparatus as claimed in claims 42, wherein the thresholding technique is one of global thresholding technique, or local thresholding technique.

48. The apparatus as claimed in claim 36, wherein the clustering technique is one of k-mean clustering technique, or fuzzy c-means clustering technique, or Bayes classifier technique, or Markov random field technique.

49. The apparatus as claimed in claim 39, wherein the clustering technique is one of k-mean clustering technique, or fuzzy c-means clustering technique, or Bayes classifier technique, or Markov random field technique.

50. The apparatus as claimed in claim 43, wherein the clustering technique is one of k-mean clustering technique, or fuzzy c-means clustering technique, or Bayes classifier technique, or Markov random field technique.

51. The apparatus as claimed in claim 37, wherein the factor analysis technique is one of principle component analysis or independent component analysis.

52. The apparatus as claimed in claim 41, wherein P is equal to six and Q is equal to ten.