Model-based Nyquist ghost correction for reverse readout echo planar imaging

Patent metadata

Field	Value
Publication number	US-12044764-B2
Application number	US-202017781660-A
Country	US
Kind code	B2
Filing date	Dec 1, 2020
Priority date	Dec 1, 2019
Publication date	Jul 23, 2024
Grant date	Jul 23, 2024

Abstract

Nyquist ghost artifacts in echo planar imaging (“EPI”) are mitigated, reduced, or otherwise eliminated by implementing robust Nyquist ghost correction (“NGC”) directly from two reversed readout EPI acquisitions. As one advantage, these techniques do not require explicit reference scanning. A model-based process is used for directly estimating statistically optimal NGC coefficients from multi-channel k-space data.

First claim

The invention claimed is:

1. A method for generating Nyquist ghost corrected magnetic resonance data, the method comprising:
(a) accessing magnetic resonance data with a computer system, the magnetic resonance data being acquired from a subject with a magnetic resonance imaging (MRI) system using a reversed readout polarity echo planar imaging (EPI) acquisition;
(b) accessing a signal model with the computer system, wherein the signal model includes a system response function that includes system response parameters associated with phase modulations corresponding to Nyquist ghost artifacts;
(c) constructing a cost function with the computer system, wherein the cost function is based on the signal model;
(d) generating estimated system response parameter values by inputting the magnetic resonance data to the cost function and minimizing the cost function with the computer system, generating output as the estimated system response parameter values; and
(e) generating Nyquist ghost corrected data with the computer system using the estimated system response parameter values.

2. A method for generating corrected magnetic resonance data, the method comprising:
(a) accessing magnetic resonance data with a computer system, the magnetic resonance data being acquired from a subject with a magnetic resonance imaging (MRI) system using a reversed readout polarity echo planar imaging (EPI) acquisition;
(b) accessing a signal model with the computer system, wherein the signal model includes a system response function that models phase modulations corresponding to one or more artifacts;
(c) constructing a cost function with the computer system, wherein the cost function is based on the signal model;
(d) generating estimated system response parameter values by inputting the magnetic resonance data to the cost function and minimizing the cost function with the computer system, generating output as the estimated system response parameter values; and
(e) generating corrected data with the computer system using the estimated system response parameter values.

3. The method of claim 1 , wherein the system response function separately models positive polarity readout gradients and negative polarity readout gradients.

4. The method of claim 1 , wherein the cost function is constructed based in part on a difference between the magnetic resonance data and the system response function applied to an estimate of the Nyquist ghost corrected data.

5. The method of claim 4 , wherein constructing the cost function comprises constructing an initial cost function and then reducing a dimensionality of the initial cost function using a variable projection of the estimate of the Nyquist ghost corrected data into the cost function.

6. The method of claim 1 , wherein minimizing the cost function includes using a gradient descent.

7. The method of claim 6 , wherein the gradient descent is a preconditioned gradient descent.

8. The method of claim 7 , wherein the preconditioned gradient descent implements a dense preconditioner.

9. The method of claim 1 , further comprising reconstructing an image from the Nyquist ghost corrected data, the image depicting the subject with significantly mitigated Nyquist ghost artifacts.

10. The method of claim 1 , wherein the reversed readout polarity EPI acquisition comprises two sequential phase-encoded EPI acquisitions whose readout gradient polarities have opposite signs at onset.

11. The method of claim 2 , wherein the system response function accounts for phase modulations that correspond to readout gradient induced eddy currents.

12. The method of claim 11 , wherein the system response function accounts for phase modulations that correspond to low-order readout gradient induced eddy currents.

13. The method of claim 2 , wherein the system response function accounts for phase modulations that correspond to group delays caused by receive chain imperfections of the MRI system used to acquire the magnetic resonance data.

14. The method of claim 2 , wherein the system response function accounts for phase modulations that correspond to Nyquist ghost artifacts.

15. The method of claim 2 , wherein the system response function in the signal model models phase modulations corresponding to anisotropic gradient delays.

16. The method of claim 2 , wherein the system response function in the signal model models phase modulations corresponding to cross-term readout gradient induced eddy currents.

17. The method of claim 2 , wherein the system response function in the signal model models phase modulations corresponding to encoding gradient induced eddy currents.

18. The method of claim 2 , further comprising reconstructing an image from the corrected data, the image depicting the subject with significantly mitigated artifacts associated with the phase modulations modeled by the signal model.

19. The method of claim 2 , wherein the reversed readout polarity EPI acquisition comprises two sequential phase-encoded EPI acquisitions whose readout gradient polarities have opposite signs at onset.

Assignees

• Mayo Found Medical Education & Res

Inventors

• Trzasko Joshua D
• Yarach Uten
• Bernstein Matthew A
• In Myung-Ho
• Sui Yi

Classifications

• G01R33/5616 Primary

  using gradient refocusing, e.g. EPI · CPC title

• G01R33/56554 Primary

  caused by acquiring plural, differently encoded echo signals after one RF excitation, e.g. correction for readout gradients of alternating polarity in EPI · CPC title