Systems and methods for automatic voxel positioning in magnetic resonance spectroscopy
Systems and methods for automated voxel positioning in magnetic resonance spectroscopy (“MRS”) are provided. In some aspects, a method includes receiving magnetic resonance (“MR”) imaging data acquired from a subject using an MR imaging system and registering the MR imaging data to an atlas having a pre-defined volume of interest (“VOI”), or segmenting a region of interest (“ROI”) directly from the MR data. The method also includes generating registration parameters based on the registration, and computing a transformed VOI using the pre-defined VOI in the atlas and the registration parameters. Alternatively, the VOI may be obtained by directly estimating it from the ROI.
1. A computer-implemented method for automated voxel positioning in magnetic resonance spectroscopy (“MRS”), the method comprising:
a) receiving magnetic resonance (“MR”) imaging data acquired from a subject using magnetic resonance imaging (MRI) system;
b) estimating an initial volume-of-interest (VOI) from the MR imaging data;
c) computing a VOI from the initial VOI by computing spatial parameters of the computed VOI based on the initial VOI;
d) operating the MRI system to acquire MRS data from the computed VOI using the spatial parameters of the computed VOI; and
e) generating a report indicative of the computed VOI based at least in part on the acquired MRS data.
2. The method of claim 1 , wherein estimating the initial VOI from the MR imaging data comprises:
providing an atlas having a pre-defined VOI contained therein;
generating registration parameters by computing a registration between the MR imaging data and the atlas; and
estimating the initial VOI by transforming the pre-defined VOI from a coordinate space associated with the atlas to a coordinate space associated with the subject by applying using the registration parameters.
3. The method of claim 2 , wherein the pre-defined VOI is associated with at least one of a white matter (“WM”) tissue, a grey matter (“GM”) tissue, a cerebrospinal fluid (“CSF”), a posterior cingulate cortex tissue, or a hippocampus tissue.
4. The method of claim 2 , wherein the registration parameters are generated by computing at least one of a linear registration, a non-linear registration, an affine registration, or a b-spline registration between the MR imaging data and the atlas.
5. The method of claim 2 , wherein during the registration the atlas is allowed to move while the MR imaging data is fixed in space.
6. The method of claim 2 , wherein during the registration the MR imaging data is allowed to move while the atlas is fixed in space.
7. The method of claim 1 , wherein estimating the initial VOI from the MR imaging data comprises computing a region-of-interest (ROI) by segmenting the MR imaging data and selecting the ROI as the initial VOI.
8. The method of claim 1 , wherein the spatial parameters of the VOI are computed by computing a minimum bounding box on the initial VOI.
9. The method of claim 8 , wherein the minimum bounding box is computed using an oriented bounding box (OBB) algorithm.
10. The method of claim 8 , wherein the spatial parameters include coordinates of the VOI in a coordinate space associated with the subject.
11. The method of claim 8 , wherein the spatial parameters include an angulation of the VOI in a coordinate space associated with the subject.
12. The method of claim 8 , wherein the spatial parameters include a size of the VOI.
13. The method of claim 1 , wherein generating the report includes analyzing the MRS data to generate at least one spectrum.
14. The method of claim 13 , wherein generating the report further comprises comparing the at least one spectrum with a reference spectrum to determine a condition of the subject.
15. A system for automated voxel positioning in magnetic resonance spectroscopy (“MRS”), the system comprising:
an input configured to receive magnetic resonance (“MR”) imaging data acquired from a subject using an MR system;
a processor programmed to:
access the MR imaging data;
generate registration parameters by registering the MR imaging data to an atlas having a pre-defined volume of interest (“VOI”);
compute a transformed VOI using the pre-defined VOI in the atlas and the registration parameters;
generate a report indicative of the transformed VOI; and
an output for displaying the report.
16. The system of claim 15 , wherein the pre-defined VOI comprises at least one of a white matter (“WM”) tissue, a grey matter (“GM”) tissue, a cerebrospinal fluid (“CSF”), a posterior cingulate cortex tissue, and a hippocampus tissue.
17. The system of claim 15 , wherein the processor is further programmed to apply at least one of a linear registration, a non-linear registration, an affine registration, and a b-spline registration.
18. The system of claim 15 , wherein the processor is further programmed to generate a minimum bounding box associated with the transformed VOI.
19. The system of claim 15 , wherein the processor is further programmed to direct the MR system to acquire MRS data based on the transformed VOI.
20. The system of claim 19 , wherein the processor is further programmed to analyze the MRS data to generate at least one spectrum.