Fix/pre free surfer pipeline #314
Conversation
…ters for brain extraction
…fering resolutions
…tments and brain extraction checks
…stortion correction
… phase encodes and patient positioning
|
Is TakJim affiliated with Takuya Hayashi's lab at RIKEN? These changes appear to potentially be related to trying to merge the NHP HCP Pipelines back into the HCP Pipelines. We should encourage this. |
| --SEPhaseNeg2=${SpinEchoPhaseEncodeNegative2} \ | ||
| --SEPhasePos2=${SpinEchoPhaseEncodePositive2} \ | ||
| --SEPhaseZero=${SpinEchoPhaseEncodeZero} \ |
There was a problem hiding this comment.
You need to include the edits to T2wToT1wDistortionCorrectAndReg.sh that add this functionality.
| --SEPhaseNeg2=${SpinEchoPhaseEncodeNegative2} \ | ||
| --SEPhasePos2=${SpinEchoPhaseEncodePositive2} \ | ||
| --SEPhaseZero=${SpinEchoPhaseEncodeZero} \ | ||
| --echospacing=${DwellTime} \ |
There was a problem hiding this comment.
DwellTime isn't defined, and also --echospacing isn't supported by the human version of T2wToT1wDistortionCorrectAndReg.sh. Is this intended to be different from --seechospacing?
There was a problem hiding this comment.
$DwellTime still isn't defined in this script.
There was a problem hiding this comment.
We should not be using DwellTime as an internal variable name for EchoSpacing. DwellTime is something different from echo spacing, and I believe that we corrected this nomenclature in PreFreeSurferPipeline quite a while ago. Also, the name of the argument in T2wToT1wDistortionCorrectAndReg isn't --echospacing, it's --seechospacing. Are you using the current version of scripts as the underlying code on which this PR is built upon?
| @@ -628,6 +870,8 @@ if [ "$CustomBrain" = "NONE" ] ; then | |||
| else # -- No T2w image | |||
|
|
|||
| log_Msg "Performing Bias Field Correction using T1w image only" | |||
| BiasFieldFastSmoothingSigma="20" | |||
| BiasFieldFastSmoothingSigma="--bfsigma=${BiasFieldFastSmoothingSigma}" | |||
There was a problem hiding this comment.
This doesn't get used, and hardcodes a smoothing kernel size.
| RefRes=$(${FSLDIR}/bin/fslval ${T1wTemplate} pixdim1 | awk '{printf "%0.2f", $1}') | ||
| log_Msg "Resolution of structure: $StrucRes" | ||
| log_Msg "Resolution of T1wTemplate: $RefRes" | ||
| if [ ! "$StrucRes" == "$RefRes" ] ; then |
There was a problem hiding this comment.
This would trigger on, say, a 0.71mm human scan (or even a 0.7mm if someone specified the 0.8mm template), and use a voxel grid different than the specified template file for all the MNINonLinear/ volume files. I'm not sure what we want to happen.
| --echospacing=${DwellTime} \ | ||
| --echospacing=${SEEchoSpacing} \ |
There was a problem hiding this comment.
Specifying the same flag twice is not correct.
Pull Request Summary
This PR refactors the processing pipeline for enhanced modularity, flexibility, and improved parameter control in T1w and T2w image processing.
Key Changes