Three-dimensional phase contrast imaging using interleaved projection data

• US 20030135103A1
• Filed: 11/08/2002
• Published: 07/17/2003
• Est. Priority Date: 11/12/2001
• Status: Active Grant

• Current Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Original Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Inventors: Mistretta, Charles A.
• Granted Patent: US 6,954,067 B2
• Time in Patent Office: Days
• US Class Current: 600/410
• CPC Class Codes: A61B 5/7289 Retrospective gating, i.e. associating measured signals or images with a physiological event after the actual measurement or image acquisition, e.g. by simultaneously recording an additional physiological signal during the measurement or image acquisition
• CPC Class Codes: A61B 5/7292 Prospective gating, i.e. predicting the occurrence of a physiological event for use as a synchronisation signal
• CPC Class Codes: G01R 33/56308 Characterization of motion or flow; Dynamic imaging
• CPC Class Codes: G01R 33/56316 involving phase contrast techniques

Abstract

A three dimensional projection reconstruction pulse sequence is employed to acquire velocity encoded NMR data from which an image indicative of spin motion is reconstructed. The velocity encoding is along all three axes and it may include acquisitions at more than one velocity encoding first moment M1. When more than one first moment M1 is acquired, a 1DFT along the velocity encoding axis is performed prior to reconstructing images from the acquired NMR data.

Claims

1. A method for producing a magnetic resonance image indicative of spin motion, the steps comprising:

a) acquiring an NMR signal with an MRI system using a three-dimensional projection reconstruction pulse sequence which includes;

producing an RF excitation field;

producing a motion encoding gradient field G_M directed along a motion encoding axis; and

acquiring an NMR signal while producing a readout gradient to sample a three-dimensional k-space volume along a trajectory that extends substantially radially from the center of k-space;

b) producing a first k-space data set by repeating step a) using different readout gradients to sample k-space throughout the three-dimensional k-space volume along different trajectories with substantially uniform spacing between trajectories;

c) producing a second k-space data set by repeating steps a), and b) using a different motion encoding gradient G_M;

d) reconstructing a first phase image from the first k-space data set;

e) reconstructing a second phase image from the second k-space data set; and

f) producing a velocity image by combining the first and second phase images.

2. The method as recited in claim 1 in which the projection angles produced while acquiring the first k-space data set are interleaved with the projection angles produced while acquiring the second k-space data set.

3. The method as recited in claim 1 in which the different motion encoding gradient G_M employed in step c) is perpendicular to the initially produced motion encoding gradient G_M and the method further includes:

g) producing a third k-space data set by repeating steps a), and b) using a different motion encoding gradient G_M which is perpendicular to the first two motion encoding gradients G_M;

h) reconstructing a third phase image from the third k-space data set; and

wherein step f) is performed by combining the first, second and third phase images.

4. The method as recited in claim 3 in which the projection angles produced while acquiring the first, second and third k-space data sets are interleaved.

5. The method as recited in claim 3 which includes:

i) producing a reference k-space data set by repeating steps a), and b) without producing a motion encoding gradient G_M;

j) reconstructing a reference phase image from the reference k-space data set; and

k) subtracting the reference phase image from each of the first, second and third phase images prior to combining them in step f).

6. The method as recited in claim 5 in which the first, second and third phase images are combined by calculating the square root of the sum of their squares of phase values at corresponding pixel locations in the first, second and third phase images.

7. The method as recited in claim 1 in which the initial motion encoding gradient field G_M has a first moment +M1 and the second motion encoding gradient field G_M has an opposite first moment −

M1 and step f) is performed by subtracting the first and second phase images.

8. The method as recited in claim 7 in which steps a)-c) are repeated a plurality of times using motion encoding gradient fields G_M having different valued first moments +M1 and −

M1 for each repeat and to thereby produce a corresponding plurality of first k-space data sets and a corresponding plurality of second k-space data sets, and wherein step f) includes subtracting corresponding k-space data sets in said plurality of first and second k-space data sets to form a plurality of subtracted k-space data sets, and performing a one-dimensional Fourier transformation along corresponding locations in the plurality of subtracted k-space data sets to produce velocity bin data sets.

9. The method as recited in claim 8 in which step f) includes selecting values from the velocity bin data sets to produce the velocity image.

21. The method as recited in claim 1 in which the sampling performed in step b) substantially undersamples the three-dimensional k-space volume.

10. A method for producing images indicative of spin motion at different phases of a cardiac cycle, the steps comprising:

a) producing a gating signal which indicates the different cardiac phases;

b) acquiring an NMR signal with an MRI system using a three-dimensional projection reconstruction pulse sequence which is performed during each of the indicated cardiac phases by;

producing an RF excitation field;

producing a motion encoding gradient field G_M directed along a first motion encoding axis; and

acquiring an NMR signal while producing a readout gradient to sample a three-dimensional k-space volume along a trajectory that extends substantially radially from the center of k-space;

c) producing a first k-space data set for each cardiac phase by repeating step b) using a different readout gradient to sample a three-dimensional k-space volume with substantially uniform spacing between trajectories; and

d) reconstructing a velocity image from each of said first k-space data sets.

11. The method as recited in claim 10 which includes:

e) producing a second k-space data set for each cardiac phase by performing steps b) and c) using a different motion encoding gradient field G_M directed along a second motion encoding axis which is perpendicular to said first motion encoding axis; and

wherein step d) includes reconstructing a second velocity image from each of said second k-space data sets.

12. The method as recited in claim 11 which includes:

f) producing a third k-space data set for each cardiac phase by performing steps b) and c) using a different motion encoding gradient field G_M directed along a third motion encoding axis which is perpendicular to said first and second motion encoding axes; and

wherein step d) includes reconstructing a third velocity image from each of said third k-space data sets.

13. The method as recited in claim 12 in which step d) includes:

d)i) regridding the first, second and third k-space data sets to form rectilinear first, second and third k-space data sets;

d)ii) Fourier transforming the first, second and third sets of rectilinear k-space data sets to form first, second and third sets of image data sets;

d)iii) calculating first, second, and third sets of velocity component images from the corresponding first, second, and third image data sets; and

d)iv) combining corresponding velocity component images in said first, second and third sets of velocity component images to form said velocity images.

14. The method as recited in claim 13 which includes:

g) producing a reference k-space data set for each cardiac phase by performing steps b) and c) with substantially no motion encoding gradient G_M;

h) regridding the reference k-space data sets;

i) Fourier transforming the regridded reference k-space data sets; and

j) calculating a set of reference phase images from the Fourier transformed reference k-space data sets; and

wherein step d)iii) includes calculating first, second and third sets of phase images from the corresponding first, second and third image data sets and subtracting respective reference phase images from corresponding phase images in said first, second and third sets of phase images.

15. The method as recited in claim 10 in which step c) is performed by repeatedly under sampling the three-dimensional k-space volume over a plurality of successive cardiac cycles, and each repeated under sampling being performed with a corresponding plurality of trajectories that are interleaved with the trajectories for the other under samplings.

16. A method for producing an image indicative of the velocity of moving spins in a subject, the steps comprising:

a) acquiring a first k-space data set with a magnetic resonance imaging system which repeatedly performs a three-dimensional projection reconstruction pulse sequence that includes;

a)i) producing an RF excitation field;

a)ii) producing a motion encoding gradient field G_M directed along a first motion encoding axis and having a first moment M1 of a first selected value;

a)iii) acquiring an NMR signal while producing a readout gradient to sample a three-dimensional k-space volume along a trajectory that extends substantially radially from the center of k-space;

a)iv) repeating steps a)i), a)ii) and a)iii) a plurality of times using different readout gradients to sample the three-dimensional k-space volume substantially uniform in all directions from the center of k-space;

b) acquiring a second k-space data set by repeating step a) using a motion encoding gradient field G_M directed along a second motion encoding axis which is perpendicular to the first motion encoding axis;

c) reconstructing a first velocity component image from the first k-space data set; and

d) reconstructing a second velocity component image from the second k-space data set.

17. The method as recited in claim 16 in which the readout gradients produced during step b) are different from those produced during step a) such that the sampling trajectories in step b) are interleaved with the sampling trajectories in step a).

18. The method as recited in claim 16 which includes:

f) acquiring a plurality of first k-space data sets by repeating step a) a corresponding plurality of times, each repeat using a motion encoding gradient field G_M having a different value for the first moment M1;

g) acquiring a plurality of second k-space data sets by repeating step b) a corresponding plurality of times, each repeat using a motion encoding gradient field G_M having a different value for the first moment M1; and

in which

step c) includes performing a one-dimensional Fourier transformation along corresponding locations in said plurality of first k-space data sets; and

step d) includes performing a one-dimensional Fourier transformation along corresponding locations in said plurality of second k-space data sets.

19. The method as recited in claim 18 in which the readout gradients produced during step b) are different from those produced during step a) such that the sampling trajectories in step b) are interleaved with the sampling trajectories in step a).

20. The method as recited in claim 18 in which the readout gradients produced during steps f) and g) are different for each motion encoding gradient field first moment value such that the sampling trajectories for the plurality of first k-space data sets are interleaved with each other and the sampling trajectories of the plurality of second k-space data sets are interleaved with each other.

22. The method as recited in claim 16 in which the sampling performed in step a)iv) substantially undersamples the three-dimensional k-space volume.