Motion correction apparatus and method

1. A medical image processing apparatus, comprising:

an image selection circuit that obtains a series of sets of image data, each set of image data representative of a region at a different time, and performs a segmentation process to segment a bone region as an anatomical feature in at least one of the sets of image data to determine a sub-region of interest comprising the anatomical feature;

a registration circuit that, for each set of image data of a plurality of the sets of image data in the series, performs a global registration process to obtain a global registration between a reference data set of the series and the set of image data; and

a registration refinement circuit that, for each set of image data of the plurality of sets of image data:

performs a local registration process, to obtain a local registration between the reference data set of the series and the set of image data within the sub-region of interest determined by the segmentation process;

obtains a combined registration based on the global registration and the local registration;

applies the combined registration to the reference data set or to the set of image data to obtain a transformed reference data set or transformed set of image data; and

performs at least one of (a), (b) and (c) to obtain a subtracted, transformed set of image data:

(a) subtracts the set of image data and the transformed reference data set;

(b) subtracts the transformed set of image data and the reference data set; and

(c) subtracts the transformed set of image data and a further reference data set.

2. An apparatus according to claim 1 , wherein the performing of the segmentation process by the image selection circuit includes comparing an intensity level of image data of the image data set or of the reference data set to a threshold and determining the sub-region of interest according to the comparison.

3. An apparatus according to claim 2 , wherein the intensity level is a CT number.

4. An apparatus according to claim 1 , wherein the performing of the segmentation process by the image selection circuit includes performing a level set segmentation process or other segmentation process.

5. An apparatus according to claim 1 , wherein the image selection circuit selects one of the sets of image data of the series to be the reference data set.

6. An apparatus according to claim 1 , wherein at least one of:

(a) the reference data set comprises pre-contrast image data representative of the region; and

(b) the plurality of sets of image data comprises contrast-enhanced image data.

7. An apparatus according to claim 1 , wherein the reference data set is the same for each of the plurality of sets of image data of the series.

8. An apparatus according to claim 1 , wherein the reference data set comprises a contrast-enhanced data set representative of the region.

9. An apparatus according to claim 1 , wherein the reference data set is the same for each of the plurality of sets of image data, such that a motion-corrected sequence of subtracted images is obtained.

10. An apparatus according to claim 1 , wherein

the series of sets of image data comprises a plurality of contrast-enhanced data sets and a non-contrast enhanced data set, and

the registration circuit subtracts the non-contrast enhanced data set, registered or otherwise, from each of the contrast-enhanced data sets, registered or otherwise.

11. An apparatus according to claim 1 , wherein the registration circuit performs a reverse transformation.

12. An apparatus according to claim 1 , wherein the performing of the local registration process comprises determining a starting point for the local registration from the global registration.

13. An apparatus according to claim 1 , wherein the registration refinement circuit determines, for the sub-region of interest, a rigid approximation or an affine approximation of the global registration, and refines that approximation to determine the local registration for the sub-region of interest.

14. An apparatus according to claim 1 , wherein

the sub-region of interest comprises, or is at least partially surrounded, by a buffer region, and

the combining by the registration refinement circuit comprises, for the buffer region, applying at least one weighting to one or both of the global registration and the local registration.

15. An apparatus according to claim 14 , wherein the weighting or at least one of the weightings varies with distance from a central area of the sub-region of interest.

16. An apparatus according to claim 1 , wherein the anatomical feature comprises bone.

17. An apparatus according to claim 1 , wherein anatomical feature comprises at least one of a sign of pathology or at least one tumour.

18. An apparatus according to claim 1 , wherein the determining of the sub-region of interest by the image selection circuit comprises determining a boundary of the sub-region of interest and filling in at least one gap within the boundary of the sub-region of interest to form a mask for selecting data from the reference data set or from the series of sets of image data.

19. An apparatus according to claim 1 , wherein the determining of the sub-region of interest by the image selection circuit comprises determining a boundary of the sub-region of interest, and dilating the boundary of the sub-region of interest to form a mask for selecting data from the reference data set or from the plurality of sets of image data.

20. An apparatus according to claim 1 , wherein the registration comprises at least one of:

a rigid registration;

an affine registration; and

a non-rigid registration.

21. An apparatus according to claim 1 , wherein the medical image processing apparatus is further configured to perform perfusion analysis on the series of subtracted, transformed set of image data.

22. A method of medical image processing, comprising:

obtaining a series of sets of image data, each set of image data representative of a region at a different time;

determining, by circuitry, a sub-region of interest comprising an anatomical feature by performing a segmentation process to segment a bone region as the anatomical feature in at least one of the sets of image data; and

for each set of image data of a plurality of the sets of image data in the series:

performing, by the circuitry, a global registration process to obtain a global registration between a reference data set of the series and the set of image data;

performing, by the circuitry, a local registration process, to obtain a local registration between the reference data set of the series and the set of image data within the sub-region of interest;

obtaining, by the circuitry, a combined registration based on the global registration and the local registration;

applying, by the circuitry, the combined registration to the reference data set or to the set of image data to obtain a transformed reference data set or transformed set of image data; and

subtracting, by the circuitry, one of (a), (b) and (c) to obtain a subtracted, transformed set of image data:

(a) the set of image data and the transformed reference data set;

(b) the transformed set of image data and the reference data set; and

(c) the transformed set of image data and a further reference data set.

23. A non-transitory computer readable medium storing computer-readable instructions that, when executed by a computer, cause the computer to perform the method according to claim 22 .