Compressed sensing reconstruction for multi-slice and multi-slab acquisitions

Patent metadata

Field	Value
Publication number	US-10663549-B2
Application number	US-201414552539-A
Country	US
Kind code	B2
Filing date	Nov 25, 2014
Priority date	Nov 25, 2014
Publication date	May 26, 2020
Grant date	May 26, 2020

Abstract

A method for acquiring a three-dimensional image volume using a magnetic resonance imaging device includes performing a multi-slice or multi-slab acquisition process to acquire a plurality of slices or three-dimensional slabs corresponding to an imaged object. Each respective slice or three-dimensional slab included in the plurality of slices or three-dimensional slabs comprises k-space data. An iterative compressed-sensing reconstruction process is applied to jointly reconstruct the plurality of three-dimensional slabs as a single consistent volume. The iterative compressed-sensing reconstruction process solves a cost function comprising a summation of individual data fidelity terms corresponding to the plurality of three-dimensional slabs.

First claim

We claim:

1. A method for acquiring a three-dimensional image volume using a magnetic resonance imaging device, the method comprising:
performing a multi-slice or multi-slab acquisition process to acquire a plurality of slices or three-dimensional slabs corresponding to an imaged object, each respective slice or three-dimensional slab included in the plurality of slices or three-dimensional slabs comprising k-space data;
applying an iterative compressed-sensing reconstruction process to jointly reconstruct the plurality of slices or three-dimensional slabs as a single consistent volume, wherein (i) the iterative compressed-sensing reconstruction process uses a combination of the k-space data corresponding to the slices or the three-dimensional slabs, and (ii) the iterative compressed-sensing reconstruction process solves a cost function comprising an inner summation and an outer summation, wherein the inner summation is a summation of individual data fidelity terms from each slice or slab in corresponding to the plurality of slices or three-dimensional slabs and the outer summation is calculated over the number of coils of the magnetic resonance imaging device; and
presenting an image depicting the single consistent volume on a display.

2. The method of claim 1 , wherein each individual data fidelity term corresponds to a comparison of k-space data associated with a respective slice or three-dimensional slab and a masked region of a current estimated image volume corresponding to the respective slice or three-dimensional slab.

3. The method of claim 2 , wherein the masked region of the current estimated image volume is a zero-padded matrix of data with non-zero data elements at positions corresponding to the respective slice or three-dimensional slab.

4. The method of claim 1 , wherein the cost function further comprises a regularization portion which is independent of a total number of slices or three-dimensional slabs.

5. The method of claim 4 , wherein the iterative compressed-sensing reconstruction process uses a proximal gradient algorithm to solve the cost function over a plurality of iterations.

6. The method of claim 4 , wherein the regularization portion applies a wavelet transform to a current estimated image volume during each iteration.

7. The method of claim 1 , wherein the multi-slice or multi-slab acquisition process is a multi-slab acquisition process which performs incoherent undersampling of the imaged object along a slab direction.

8. The method of claim 1 , wherein the multi-slab acquisition process employs an accelerated 3D Time-of-Flight (TOF) sampling strategy.

9. The method of claim 8 , wherein the accelerated 3D TOF sampling strategy employs variable density spiral phyllotaxis, or variable density Poisson patterns in ky-kz phase-encoding directions.

10. An article of manufacture for acquiring a three-dimensional image volume using a magnetic resonance imaging device, the article of manufacture comprising a non-transitory, tangible computer-readable medium holding computer-executable instructions for performing a method comprising:
using a plurality of coils included in the magnetic resonance imaging device to perform an undersampled multi-slab scan of a region of interest to yield a plurality of three-dimensional slabs;
jointly reconstructing the plurality of three-dimensional slabs as a single consistent volume by solving a cost function comprising an inner summation and an outer summation, wherein the inner summation is a summation of individual data fidelity terms from each slab in corresponding to the plurality of three-dimensional slabs, wherein the cost function uses a combination of k-space corresponding to the three-dimensional slices acquired during the undersampled multi-slab scan and the outer summation is calculated over the number of coils of the magnetic resonance imaging device; and
presenting an image depicting the single consistent volume on a display.

11. The article of manufacture of claim 10 , wherein each individual data fidelity term corresponds to a comparison of k-space data associated with a respective three-dimensional slab and a masked region of a current estimated image volume corresponding to the respective three-dimensional slab.

12. The article of manufacture of claim 11 , wherein the masked region of the current estimated image volume is a zero-padded matrix of data with non-zero data elements at positions corresponding to the respective three-dimensional slab.

13. The article of manufacture of claim 10 , wherein the cost function further comprises a regularization portion which is independent of a total number of three-dimensional slabs.

14. The article of manufacture of claim 13 , wherein the cost function is solved over a plurality of iterations using a proximal gradient algorithm.

15. The article of manufacture of claim 13 , wherein the regularization portion applies a wavelet transform to a current estimated image volume during each iteration.

16. The article of manufacture of claim 10 , wherein the plurality of three-dimensional slabs are acquired using incoherent undersampling of the region of interest along a slab direction.

17. The article of manufacture of claim 10 , wherein the plurality of three-dimensional slabs are acquired using an accelerated 3D Time-of-Flight (TOF) sampling strategy.

18. The article of manufacture of claim 17 , wherein the accelerated 3D TOF sampling strategy employs variable density spiral phyllotaxis trajectory in ky-kz phase-encoding directions.

19. A system for performing multi-slab acquisitions with compressed sensing reconstruction, the system comprising:
a display;
an imaging device comprising a plurality of coils configured to acquire a plurality of three-dimensional slabs corresponding to an imaged object, each respective three-dimensional slab included in the plurality of three-dimensional slabs comprising k-space data;
a central control computer unit configured to:
apply an iterative compressed-sensing reconstruction process to the plurality of three-dimensional slabs to yield an image volume, wherein (i) the iterative compressed-sensing reconstruction process uses a combination of the k-space data corresponding to the three-dimensional slabs, and (ii) the iterative compressed-sensing reconstruction process solves a cost function comprising an inner summation and an outer summation, wherein the inner summation is a summation of individual data fidelity terms from each slab in corresponding to the plurality of three-dimensional slabs and the outer summation is calculated over the a number of coils of the magnetic resonance imaging device; and
presenting an image depicting the single consistent volume on the display.

20. The system of claim 19 , wherein the central control computer unit is further configured to:
use a proximal gradient algorithm during the iterative compressed-sensing reconstruction process to solve the cost function over a plurality of iterations.

Assignees

• Siemens Healthcare Gmbh

Inventors

• Mailhe Boris
• Nadar Mariappan S
• Stalder Aurélien
• Wang Qiu
• Zenge Michael

Classifications

• G01R33/4835

  of multiple slices · CPC title

• G01R33/5608 Primary

  Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels (image data processing or generation, in general G06T) · CPC title

• G01R33/4818

  MR characterised by data acquisition along a specific k-space trajectory or by the temporal order of k-space coverage, e.g. centric or segmented coverage of k-space · CPC title

• G01R33/5611 Primary

  Parallel magnetic resonance imaging, e.g. sensitivity encoding [SENSE], simultaneous acquisition of spatial harmonics [SMASH], unaliasing by Fourier encoding of the overlaps using the temporal dimension [UNFOLD], k-t-broad-use linear acquisition speed-up technique [k-t-BLAST], k-t-SENSE (structural details of arrays of sub-coils G01R33/3415) · CPC title