Difference between revisions of "Inter-slice Motion Correction for fMRI"
From NAMIC Wiki
Rmachiraju (talk | contribs) (→Images) |
Rmachiraju (talk | contribs) (→Images) |
||
Line 23: | Line 23: | ||
[[Image:original.jpg]] | [[Image:original.jpg]] | ||
− | Volume rendering of original volume | + | '''Volume rendering of original volume''' |
Line 33: | Line 33: | ||
[[Image:firdaus_affine_rms.png]] | [[Image:firdaus_affine_rms.png]] | ||
− | + | Volume rendering of affine motion with mean square error metric | |
[[Image:firdaus_affine_mi.png]] | [[Image:firdaus_affine_mi.png]] |
Revision as of 17:34, 12 September 2007
Home < Inter-slice Motion Correction for fMRIContents
Objective
* To perform inter-slice motion correction in fMR images using affine and non-rigid registration methods.
Progress
* Implemented and tested b-Spline/affine registration with Mattes-MI and KL metrics. * Coding completed. However, results are far from satisfactory. * Progress (maybe) stymied by "poor" resolution of data. * Progress (maybe) stymied by "not-so-viable" metrics (MI, etc.).
Results
Affine
- affine motion with all metrics degrades the inter-slice registration.
Non-Rigid
- B-spline deformation with KL metric produces very minimal change (in a least squares sense), while MI produces more changes but doesn't look correct.
Images
Volume rendering of original volume
Volume rendering of up-sampled volume (0.5x0.5mm in axial plane)
Volume rendering of co-registered volume
Volume rendering of affine motion with mean square error metric
Affine motion with Mattes MI error metric
Issues
- As can be seen any registration result looks more jagged (along the sagittal and coronal sections).
- Low sampling rate and image noise - all the metrics are extremely non-smooth as a result.
- Even the slightest change in initialization results in a large deviation in the registration result.
Open questions
* Intensity normalization / histogram equalization - would these have any impact on inter-slice registration? Would they affect the computation of the joint-entropies for MI/KL metrics. * Should there be some regularization factor on the registration to limit the motion ?
To Do
* Right now, registration is done on a slice by slice basis. However, due to the low resolution of fMRI and the absence of gradients result in low registration accuracy. We are investigating using alternative image-to-image metrics like the KL divergence. We are also looking at simultaneously registering multiple slices.
Key Investigators
* Firdaus Janoos, Raghu Machiraju, Steve Pieper, Wendy Plesniak.