-
Notifications
You must be signed in to change notification settings - Fork 10
Structural processing
Chris Klink edited this page May 26, 2020
·
5 revisions
Here we provide an overview of the main steps involved, and the tools available in PRIME-RE or in existing neuroimaging software packages, for the processing of NHP structural images with the goal of achieving an extracted brain mask and the segmentation masks (WM,GM and CSF). We also provide a list of existing pipelines for macaque anatomical processing for reference.
| Processing Step | Available Tools |
|---|---|
| 1. Data Preparation | |
| Reorientation | FSL: fslreorient2std, fslswapdim + fslreorientFreesurfer: mri_convert -sphinx, mri_convert --in_orientationJip analysis toolkit Web-based Reorient Tool |
| Deoblique | AFNI: 3drefit -deoblique (for changing header information) |
| Cropping | FSL: fslroi, FSLeyesAFNI: @clip_volumeFreeSurfer: mri_convert --slice-crop
|
| Denoising | Adaptive non-local means filter denoising in ANTs (ImageDenoise), SPM or Matlab package
|
| Averaging multiple images | Linear Registration tools: FSL-FLIRT, AFNI-3dVolReg, 3dAllineate, SPM Register, etc.Image averaging: fslmaths, SPM Imcalc, etc. |
| 2. Bias-Correction | |
| T1xT2 bias field correction (HCP Method) | Can be implemented using standard image calculation software such as fslmaths based on procedures described in Rilling et al. (2011)A module for this bias-correction is also available in Macapype ( correct_bias.py). |
N3, N4BiasFieldCorrection
|
Available in ANTs, MINC, Freesurfer packages. One could also consider N3biascorrection which works better in some cases. |
| FSL-Fast | FSL |
| CMTK-mrbias | Find it here |
| 3. Brain Extraction | |
| Template-based | AntsBrainExtraction (ANTs), Atlasbrex |
| Non Template-based | FSL-BET (can also be used with a template), bet_macaque.sh |
| Deep Learning Model | U-NET |
| Manual corrections | ITK-SNAP, Slicer, BrainBox |
| 4. Brain Segmentation | |
| Template-based | AntsAtroposN4 script, Atropos (ANTs), SPM Segment |
| Non Template-based | FSL-Fast (can be used with templates) |
| Manual segmentations/corrections | ITK-SNAP, BrainBox |
| 5. Templates and Atlases | See PRIME-RE |
| 6. Ready-to-use Pipelines | |
| Civet-Macaque | Find it here |
| NHP-Freesurfer | Find it here |
| PREEMACS | Find it here |
| Macapype | Find it here |
| Precon_all | Find it here |
| HCP-style NHP Pipeline | Find it here |
A. Why the interest in NHP neuroimaging?
B. What makes NHP MRI challenging?
C. Typical data analysis challenges
D. Structural data processing steps and PRIME-RE tools
E. Functional data processing steps and PRIME-RE tools
F. Diffusion data processing steps and PRIME-RE tools
G. Cross-species comparisons and PRIME-RE tools