Yes I apologize for coding the bias field wrong in the initial versions of the HCP Pipeline, but the SEBASED is much better.
Peace, Matt. From: "Sanchez, Juan (NYSPI)" <juan.sanc...@nyspi.columbia.edu<mailto:juan.sanc...@nyspi.columbia.edu>> Date: Thursday, March 29, 2018 at 12:36 PM To: Matt Glasser <glass...@wustl.edu<mailto:glass...@wustl.edu>>, Keith Jamison <kjami...@umn.edu<mailto:kjami...@umn.edu>> Cc: "hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>" <hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>> Subject: Re: [HCP-Users] Intensity Normalization 3_22 Thanks Matt it does look good. I will switch to LEGACY from SEBASED thanks ________________________________ From: Glasser, Matthew <glass...@wustl.edu<mailto:glass...@wustl.edu>> Sent: Thursday, March 29, 2018 1:35:28 PM To: Sanchez, Juan (NYSPI); Keith Jamison Cc: hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] Intensity Normalization 3_22 ATTENTION: This email came from an external source. Do not open attachments or click on links from unknown senders or unexpected emails. That image looks okay, what is the issue? Peace, Matt. From: "Sanchez, Juan (NYSPI)" <juan.sanc...@nyspi.columbia.edu<mailto:juan.sanc...@nyspi.columbia.edu>> Date: Thursday, March 29, 2018 at 12:29 PM To: Keith Jamison <kjami...@umn.edu<mailto:kjami...@umn.edu>>, Matt Glasser <glass...@wustl.edu<mailto:glass...@wustl.edu>> Cc: "hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>" <hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>> Subject: Re: [HCP-Users] Intensity Normalization 3_22 Thanks Keith We are uing SEBASED as we are using TOPUP dc. Thanks for catching that. I reran with: "div BiaseField2" and got the attached results fslmaths Task_fMRI_emomo_1_nonlin.nii.gz -div BiasField.2.nii.gz -mul Jacobian_MNI.2.nii.gz -mas Task_fMRI_emomo_1_nonlin_mask.nii.gz -thr 0 -ing 1000 output_BiasField2 -odt float the_sebased_bias fslmaths Task_fMRI_emomo_1_nonlin.nii.gz -div Task_fMRI_emomo_1_sebased_bias.nii.gz -mul Jacobian_MNI.2.nii.gz -mas Task_fMRI_emomo_1_nonlin_mask.nii.gz -thr 0 -ing 1000 output -odt float and replicated the original error. ________________________________ From: Keith Jamison <kjami...@umn.edu<mailto:kjami...@umn.edu>> Sent: Thursday, March 29, 2018 12:35:38 PM To: Glasser, Matthew Cc: Sanchez, Juan (NYSPI); hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] Intensity Normalization 3_22 ATTENTION: This email came from an external source. Do not open attachments or click on links from unknown senders or unexpected emails. The command you ran locally is using the Jacobian as the bias field, which is incorrect, and the "-div Jacobian -mul Jacobian" is just cancelling out any effect (output has same bias as input fMRI). It should instead be "-div BiasField.2 -mul Jacobian_MNI.2". That said, your original output looks like the bias field was incorrectly estimated. Did you use --biascorrection=SEBASED in your call to GenericfMRIVolumeProcessingPipeline? If you used --biascorrection=LEGACY (might be the default?) you may also want to check if your MNINonLinear/T1w_restore and MNINonLinear/T2w_restore look properly bias corrected. -Keith On Thu, Mar 29, 2018 at 11:55 AM, Glasser, Matthew <glass...@wustl.edu<mailto:glass...@wustl.edu>> wrote: I think that might be an old version of the pipelines. If you run on the latest version is it better? Peace, Matt. From: <hcp-users-boun...@humanconnectome.org<mailto:hcp-users-boun...@humanconnectome.org>> on behalf of "Sanchez, Juan (NYSPI)" <juan.sanc...@nyspi.columbia.edu<mailto:juan.sanc...@nyspi.columbia.edu>> Date: Thursday, March 29, 2018 at 10:23 AM To: "hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>" <hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>> Subject: [HCP-Users] Intensity Normalization 3_22 Dear all We are using the 3_22 Pipelines to process out data. We noticed that the processed fMRI results had an unusual intensity inhomogeneity for ALL of our runs. (first attachment) We found that the error occurued during intensity normalization Here: ${FSLDIR}/bin/fslmaths ${InputfMRI} $biascom $jacobiancom -mas ${BrainMask} -mas ${InputfMRI}_mask -thr 0 -ing 10000 ${OutputfMRI} -odt float I copied the relevant files and ran the fslmaths command localy (InputfMRI = Task_fMRI_emomo_1) fslmaths Task_fMRI_emomo_1_nonlin.nii.gz -div Jacobian_MNI.2.nii.gz -mul Jacobian_MNI.2.nii.gz -mas Task_fMRI_emomo_1_nonlin_mask.nii.gz -thr 0 -ing 1000 output -odt float The output (second attachment) looked correct. I have tried to replicate the error and have not been able to Can anyone suggest a possible explenation? Thanks J _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org<mailto:HCP-Users@humanconnectome.org> http://lists.humanconnectome.org/mailman/listinfo/hcp-users _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org<mailto:HCP-Users@humanconnectome.org> http://lists.humanconnectome.org/mailman/listinfo/hcp-users _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users