Method and apparatus for extended phase correction in phase sensitive magnetic resonance imaging

Patent metadata

Field	Value
Publication number	US-9449386-B2
Application number	US-201314380972-A
Country	US
Kind code	B2
Filing date	Feb 27, 2013
Priority date	Feb 28, 2012
Publication date	Sep 20, 2016
Grant date	Sep 20, 2016

Abstract

Methods, apparatuses, systems, and software for extended phase correction in phase sensitive Magnetic Resonance Imaging. A magnetic resonance image or images may be loaded into a memory. Two vector images A and B associated with the loaded image or images may be calculated either explicitly or implicitly so that a vector orientation by one of the two vector images at a pixel is substantially determined by a background or error phase at the pixel, and the vector orientation at the pixel by the other vector image is substantially different from that determined by the background or error phase at the pixel. A sequenced region growing phase correction algorithm may be applied to the vector images A and B to construct a new vector image V so that a vector orientation of V at each pixel is substantially determined by the background or error phase at the pixel. A phase corrected magnetic resonance image or images may be generated using the vector image V, and the phase corrected magnetic resonance image or images may be displayed or archived.

First claim

The invention claimed is:

1. A computerized method for generating a phase corrected magnetic resonance image or images comprising:
(a) acquiring a magnetic resonance image or images containing background or error phase information;
(b) calculating two vector images A and B using the acquired image or images so that a vector orientation by one of the two vector images at a pixel is determined by the background or error phase at the pixel, and the vector orientation at the pixel by the other vector image is different from that determined by the background or error phase at the pixel;
(c) applying a sequenced region growing phase correction algorithm to the vector images A and B to construct a new vector image V, wherein the algorithm comprises:
(i) selecting an initial seed pixel or pixels and assigning either A or B of the initial seed pixel or pixels as a value of V for the initial seed pixel or pixels;
(ii) selecting a secondary seed pixel and selecting either A or B of the secondary seed pixel as a value of V for the secondary seed pixel based on whether A or B of the secondary seed pixel has a smaller angular difference with an estimated V for the secondary seed pixel;
(iii) determining for the secondary seed pixel a local quality metric for each of the nearest neighbor pixels of the secondary seed pixel for which V has not been determined and assigning a priority to each of the nearest neighbor pixels using the local quality metric in order to determine the sequence by which each of the nearest neighbor pixels is to be selected as a further seed pixel;
(iv) repeating the steps of (ii) and (iii) to complete the sequenced region growing with respect to further seed pixels and to construct the vector image V so that a vector orientation of V at each pixel is substantially determined by the background or error phase at the pixel;

(d) generating the phase corrected magnetic resonance image or images from the acquired magnetic resonance image or images using the vector image V; and
(e) displaying or archiving the phase corrected magnetic resonance image or images.

2. The method of claim 1 , wherein an initial seed pixel or pixels are placed onto a high priority pixel stack or stacks among a series of pixel stacks that are initially empty and which facilitate a sequencing of the sequenced region growing.

3. The method of claim 1 , wherein a pixel is selected as a secondary seed pixel if it has not been processed previously as a seed pixel and it is on a pixel stack that has a highest priority among pixel stacks that contain at least one pixel that has not been processed as a seed pixel.

4. The method of claim 1 , wherein the local quality metric of a pixel is calculated as the smaller of two orientational differences between A and B of the pixel with an estimated V for the pixel.

5. The method of claim 4 , wherein the estimated V for a pixel is a zeroth order estimation calculated as an average of V for pixels located within a neighboring region of the pixel and for which V has been previously determined.

6. The method of claim 5 , wherein a size of the neighboring region is either fixed or adaptively adjusted based on a local quality metric for the pixel.

7. The method of claim 1 , wherein a maximum possible range of 0-π for the angular difference between any two vectors is used to gauge and bin the local quality metric and to place a pixel onto a pixel stack and wherein the pixel stack covering a subrange of 0-πfor the quality metric is assigned a priority, and wherein a pixel stack of a higher priority is for receiving pixels with a smaller quality metric and a pixel stack of a lower priority is for receiving pixels with a larger quality metric.

8. The method of claim 7 , wherein the priority of a pixel stack from which a pixel is selected as a seed pixel is recorded for the sequenced region growing as a quality metric index to reflect an integrity of the sequenced region growing.

9. The method of claim 8 , wherein the quality metric index is used to segment an image into different segments of possible inconsistent region growing and then to combine the different segments into an overall consistent region growing to form a final vector image V.

10. The method of claim 1 , wherein a value of the vector A for an initial seed pixel is assigned as V A , and a sequenced region growing is performed to construct a vector image V A , and wherein a value of the vector B for the same initial seed pixel is assigned as V B , and another sequenced region growing is performed to construct a vector image V B .

11. The method of claim 10 , wherein either vector image V A or vector image V B is set to be a final vector image V, depending on whether vector image V A or vector image V B has a greater overall orientational smoothness.

12. The method of claim 1 , wherein the sequenced region growing is performed in two or three spatial dimensions or by including the temporal dimension for a series of dynamically acquired images.

13. The method of claim 1 wherein acquiring a magnetic resonance image or images comprises acquiring two-point Dixon water and fat images, wherein a first image S1 is acquired at a first echo time TE1 and a second image S2 is acquired at a second echo time TE2.

14. The method of claim 13 , wherein the images S i and S 2 are expressed according to the following equations:
S 1 =( W+δ 1 Fe iθ 1 ) P 1
S 2 =( W+δ 2 Fe iθ 2 ) P 1 P
where W and F are amplitudes for water and fat respectively, P 1 is a phase factor of image S 1 , P is an additional phase factor of image S 2 relative to image S 1 and is determined by a background or error phase, and the method further comprises calculating an amplitude attenuation factor (δ 1 , δ 2 ) and phase (θ 1 , θ 2 ) as a function of two echo times (TE1, TE2) for the fat signal using a pre-determined fat spectrum.

15. The method of claim 14 , wherein the images S 1 and S 2 are used to generate two vector images A and B as expressed according to the following equations:
A=S 1 *S 2 [Q A +δ 1 (1− Q A ) e iθ 1 ][Q A +δ 2 (1− Q A ) e −iθ 2 ]

B=S 1 *S 2 [Q B +δ 1 (1− Q B ) e iθ 1 ][Q B +δ 2 (1− Q B ) e −iθ 2 ]

where Q A and Q B are the two mathematically possible solutions of the following quadratic equation of Q, which is defined as

Q
=

W

W
+
F

(i.e., the water fraction for given pixel):
[(1+δ 2 2 −2δ 2 cos θ 2 ) M 1 −(1+δ 1 2 −2δ 1 cos θ 1 ) M 2 ]Q 2 −2[(δ 2 2 −δ 2 cos θ 2 ) M 1 −(δ 1 2 −δ 1 cos θ 1 ) M 2 ]Q +[( M 1 δ 2 2 −M 2 δ 1 2 )]=0

where M 1 and M 2 are the square of the amplitudes of the images S 1 and S 2 , respectively (i.e., M 1 =|S 1 | 2 and M 2 =|S 2 | 2 .

16. The method of claim 15 , wherein the vector image V is used to phase correct and remove the phase factor P from the image S 2 , the phase corrected S 2 is combined with S 1 to solve for WP 1 and FP 1 , and then to generate a water-only image and a fat-only image according to the following equations:
W =Real{( WP 1 ) WP 1 */| WP 1 |}
F =Real{( FP 1 ) FP 1 */| FP 1 |}
where Real{. . . } is to take the real component of its complex argument, * is to take the complex conjugate of its argument, and WP 1

Assignees

• Univ Texas

Inventors

• Ma Jingfei

Classifications

• G01R33/565 Primary

  Correction of image distortions, e.g. due to magnetic field inhomogeneities · CPC title

• G06T12/30

  Image post-processing, e.g. metal artefact correction · CPC title

• G06F18/24147

  Distances to closest patterns, e.g. nearest neighbour classification · CPC title

• G06F18/2113

  by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title