Method for reconstructing motion-compensated magnetic resonance images from non-Cartesian k-space data
A method for reconstructing a motion-compensated image depicting a subject with a magnetic resonance imaging (MRI) system is provided. An MRI system is used to acquire a time series of k-space data from the subject by sampling k-space along non-Cartesian trajectories, such as radial, spiral, or other trajectories at a plurality of time frames. Those time frames at which motion occurred are identified and this information used to segment the time series into a plurality of k-space data subsets. For example, the k-space data subsets contain k-space data acquired at temporally adjacent time frames that occur between those identified time frames at which motion occurred. Motion correction parameters are determined from the k-space data subsets. Using the determined motion correction parameters, the k-space data subsets are corrected for motion. The corrected data subsets are combined to form a corrected k-space data set, from which a motion-compensated image is reconstructed.
1. A method for reconstructing a motion-compensated image depicting a subject with a magnetic resonance imaging (MRI) system, the steps of the method comprising:
a) acquiring with the MRI system, a time series of k-space data sets by sampling k-space along non-Cartesian trajectories at a plurality of time frames;
b) identifying in the time series of k-space data sets acquired in step a), ones of the plurality of time frames at which motion occurred;
c) segmenting the time series of k-space data acquired in step a) to produce a plurality of k-space data subsets, each k-space data subsets containing k-space data acquired in temporally adjacent time frames occurring between the ones of the plurality of times frames at which motion occurred identified in step b);
d) determining motion correction parameters from the k-space data subsets produced in step c);
e) producing corrected k-space data subsets by applying the motion correction parameters determined in step d) to the k-space data subsets produced in step c);
f) combining the corrected k-space data subsets produced in step e) to produce a corrected k-space data set; and
g) reconstructing a motion-compensated image from the corrected k-space data set produced in step f).
2. The method as recited in claim 1 in which step f) includes identifying selected ones of the corrected k-space data subsets that contain less k-space data than a preselected threshold amount of k-space data; and in which step g) includes reconstructing a motion-compensated image from the corrected k-space data set produced in step f) using an iterative reconstruction technique in which k-space data associated with the identified selected ones of the corrected k-space data subsets are weighted by a weighting factor.
3. The method as recited in claim 2 in which the weighting factor decreases a relative contribution of the selected ones of the corrected k-space data subsets identified in step f).
4. The method as recited in claim 1 in which the non-Cartesian trajectories are at least one of radial trajectories, k-space blade trajectories, and spiral trajectories.
5. The method as recited in claim 1 in which step d) includes reconstructing subset images from the k-space data subsets produced in step c) and determining the motion correction parameters from the reconstructed subset images.
6. The method as recited in claim 5 in which step d) includes determining the motion correction parameters by:
producing an average subset image by averaging the subset images; and
co-registering each subset image to the produced average subset image.
7. The method as recited in claim 1 in which step d) includes determining the motion correction parameters by:
identifying the k-space data subset produced in step c) that contains the largest amount of data in the plurality of k-space data subsets produced in step c);
estimating an image from the identified k-space data subset; and
co-registering each subset image to the estimated image.
8. The method as recited in claim 5 in which step d) includes reconstructing each subset image from a portion of the respective consistent data subset that includes k-space data sampled from a central portion of k-space within a Nyquist radius.
9. The method as recited in claim 1 in which step b) includes calculating a center-of-mass value for k-space data in each of the plurality of time frames and comparing center-of-mass values between ones of the plurality of time frames.
10. The method as recited in claim 9 in which step b) includes forming for each of the plurality of time frames, a data vector having a length equal to a number of receiver coils used to acquire the time series of k-space data in step a) times a dimensionality of the times series of k-space data acquired in step a), the data vector containing center-of-mass values for each of the plurality of time frames.
11. The method as recited in claim 10 in which step b) includes calculating correlation values between each data vector and identifying one of the plurality of time frames as a time frame at which motion occurred when the correlation value associated with the respective data vector is above a threshold value.
12. The method as recited in claim 11 in which the threshold value is determined by a noise level in the time series of k-space data acquired in step a).
13. The method as recited in claim 1 in which the motion correction parameters are determined in step d) by co-registering the k-space data subsets.
14. The method as recited in claim 1 in which the ones of the plurality of time frames at which motion occurred are identified in step c) by:
i) calculating a data consistency measure for each time frame in the plurality of time frames; and
ii) comparing the data consistency measures for ones of the plurality of time frames.
15. The method as recited in claim 14 in which the data consistency measure is a center-of-mass value.
16. The method as recited in claim 15 in which the plurality of k-space data subsets are produced in step c) by grouping those time frames in the plurality of time frames having similar center-of-mass values.
17. The method as recited in claim 1 in which step e) includes applying the motion correction parameters determined in step d) to respective ones of the k-space data subsets produced in step c) when the respective motion correction parameters exceed a threshold value.
18. The method as recited in claim 17 in which step b) includes generating information about whether motion is detected in each of the plurality of time frames in the time series of k-space data; and in which step e) includes applying the motion correction parameters determined in step d) to respective ones of the k-space data subsets produced in step c) using the information about whether motion is detected in each of the plurality of time frames in the time series of k-space data generated in step b).
19. A method for reconstructing a motion-compensated image depicting a subject with a magnetic resonance imaging (MRI) system, the steps of the method comprising:
a) acquiring with the MRI system, a time series of k-space data sets by sampling k-space along non-Cartesian trajectories at a plurality of time frames;
b) segmenting the time series of k-space data acquired in step a) to produce a plurality of k-space data subsets using a preselected segmentation scheme in which each k-space data subset contains k-space data acquired in temporally adjacent time frames;
c) determining motion correction parameters from the k-space data subsets produced in step b);
d) producing corrected k-space data subsets by applying the motion correction parameters determined in step c) to the k-space data subsets produced in step b);
e) combining the corrected k-space data subsets produced in step d) to produce a corrected k-space data set; and
f) reconstructing a motion-compensated image from the corrected k-space data set produced in step e).
20. The method as recited in claim 19 in which the non-Cartesian trajectories are at least one of radial trajectories, k-space blade trajectories, and spiral trajectories.