Hi,

Hi Samir,

To add lesion masks to your preprocessing pipeline, the steps are the following:

1) In Setup.ROIs define a new ROI (e.g. named "lesion_mask") an enter your lesion masks (remember to check the option 'subject-specific ROI' to indicate that you have a different mask per subject)

2) when preprocesing, instead of using the default pipeline, click instead on 'load' and load the pipeline file .../conn/utils/preprocessingpipelines/defaultMNI_withlesion.mat. When running preprocessing you will be prompted to select ROIs for functional masking. Select there the option that reads "lesion_mask (exclusive mask; exclude all voxels within this ROI)". You will also be prompted to confirm normalization options, make sure there that the option that reads "use structural lesion masks: lesion_mask" is selected.

This pipeline will incorporate the lesion masks into subject-specific TPM files characterizing the lesion as a separate tissue-class during MNI-normalization, and it will mask-out functional data within the lesioned areas (feel free to modify this pipeline to adapt it to your own data or needs)

Let me know if you have any questions

Alfonso

Originally posted by Mohammad Samim Zaheer:

Hi,

Could you please clarify a couple of questions for me?

1. What are the step-by-step instructions for adding a lesion mask into CONN to exclude the lesioned areas from analysis? 

2. How should I modify the default preprocessing pipeline for subjects with lesions? ( I selected the functional indirect segmentation and normalization with lesion mask and structural segmentation and normalization with lesion mask, and kept the others the same with their respective order)

 

3. Are there any addtional modfication for subjetcs with lesion masks in next steps/tabs?

 

Now I don't see any results in the 2nd-level analysis tab for any seed/network when doing between-subjects contrast in comparing the patient with the normal population.

Thank you for your help.
Samim Zaheer