Method and system for off-resonance correction for non-cartesian parallel image reconstruction
In MR imaging using a multi-coil device, multifrequency interpolation is combined with a convolution such as BOSCO to achieve simultaneous parallel reconstruction and off-resonance correction. The unaliased and deblurred image is calculated for each coil, and the final image is formed from the images for all of the coils.
1. A method for imaging an object, the method comprising:
(a) scanning the object with a scanner having a plurality of coils to obtain a raw time signal from each of the coils;
(b) calculating an unaliased and deblurred image for each of the coils; and
(c) combining the unaliased and deblurred images into a final image;
wherein step (b) comprises: (i) demodulating the raw time signal from each of the coils to obtain a demodulated data set corresponding to each of the coils; (ii) convolving each said demodulated data set onto a Cartesian grid to form a gridded data set; (iii) forming unaliased Cartesian k-space data by performing an operation on the gridded data sets formed in step (b)(ii); and (iv) forming unaliased channel images from the unaliased Cartesian k-space data.
2. The method of claim 1 , wherein the object is scanned through a spiral scan.
3. The method of claim 1 , wherein step (b)(iii) is performed using a convolution.
4. The method of claim 3 , wherein step (b)(iii) is performed using a BOSCO convolution.
5. The method of claim 4 , wherein step (b)(iv) is performed using a Fourier transform.
6. The method of claim 3 , wherein step (b) is performed using multifrequency interpolation.
7. The method of claim 3 , wherein step (b) is performed using conjugate phase reconstruction.
8. The method of claim 3 , wherein step (b) is performed using a fast approximation to conjugate phase reconstruction.
9. The method of claim 8 , wherein the fast approximation to conjugate phase reconstruction is multifrequency reconstruction.
10. The method of claim 8 , wherein a field map is used for deblurring, and wherein the field map used for deblurring is derived using a semi-automatic method.
11. The method of claim 2 , wherein step (b) comprises:
(i) performing a reconstruction on the raw time signal from each of the coils to obtain unaliased k-space data on Cartesian grids;
(ii) generating a set of phase masks on the Cartesian grids using constant demodulation frequencies;
(iii) multiplying the unaliased k-space data with the phase masks; and
(iv) forming a set of base images from the unaliased k-space data multiplied with the phase masks.
12. The method of claim 11 , wherein step (b)(i) is performed using a BOSCO reconstruction.
13. The method of claim 11 , wherein step (b)(iv) is performed using a Fourier transform.
14. The method of claim 11 , wherein step (c) comprises combining the base images through multifrequency interpolation.
15. A system for imaging an object, the system comprising:
a scanner having a plurality of coils to obtain a raw time signal from each of the coils;
a processor, in communication with the scanner, for calculating an unaliased and deblurred image for each of the coils and combining the unaliased and deblurred images into a final image; and
an output, in communication with the processor, for outputting the final image; wherein the unaliased and deblurred image for each of the coils is calculated by:
(i) demodulating the raw time signal from each of the coils to obtain a demodulated data set corresponding to each of the coils;
(ii) convolving each said demodulated data set onto a Cartesian grid to form a gridded data set;
(iii) forming unaliased Cartesian k-space data by performing an operation on the gridded data sets formed in step (ii); and
(iv) forming unaliased channel images from the unaliased Cartesian k-space data.
16. The system of claim 15 , wherein the scanner performs a spiral scan.
17. The system of claim 15 , wherein the processor performs step (iii) using a convolution.
18. The system of claim 17 , wherein the processor perform step (iii) using a BOSCO convolution.
19. The system of claim 18 , wherein the processor performs step (iv) using a Fourier transform.
20. The system of claim 17 , wherein the processor deblurs the images using multifrequency interpolation.
21. The system of claim 17 , wherein the processor deblurs the images using conjugate phase reconstruction.
22. The system of claim 17 , wherein the processor deblurs the images using a fast approximation to conjugate phase reconstruction.
23. The system of claim 22 , wherein the fast approximation to conjugate phase reconstruction is multifrequency reconstruction.
24. The system of claim 22 , wherein a field map is used for deblurring, and wherein the field map used for deblurring is derived using a semi-automatic method.
25. The system of claim 16 , wherein the unaliased and deblurred image for each of the coils is calculated by:
(i) performing a reconstruction on the raw time signal from each of the coils to obtain unaliased k-space data on Cartesian grids;
(ii) generating a set of phase masks on the Cartesian grids using constant demodulation frequencies;
(iii) multiplying the unaliased k-space data with the phase masks; and
(iv) forming a set of base images from the unaliased k-space data multiplied with the phase masks.
26. The system of claim 21 , wherein the processor performs step (i) using a BOSCO reconstruction.
27. The system of claim 25 , wherein the processor performs step (iv) using a Fourier transform.
28. The system of claim 25 , wherein the processor combines the base images through multifrequency interpolation.