help > preprocessing questions
Feb 24, 2021  12:02 AM | ela
preprocessing questions
Hi Alfonso,

I am doing resting-state analysis with CONN and have come up with some issues. In our dataset, there are some subjects with different image parameters compared with the rest of the group. These include differences in slice numbers, TR, and one/two sessions.

1) is it still possible to keep these subjects? regarding slice number, as there is no field in conn to enter slice numbers I assume it is counted internally by the toolbox, right? for TR, I read it in this thread that I can enter different repetition times in a vector. what if the TR varies between two sessions within a subject? Also, how does conn handle single-session data in a group of two sessions?

2) I plan to do segmentation and normalization using the dartel toolbox due to the presence of abnormalities in the dataset. The steps are:
-Functional realignment and unwarp (CONN)
Slice-Timing Correction (CONN)
Outlier identification (CONN)
segmentation and normalization (DARTEL)
does this order of preprocessing seem rational?

After the DARTEL step, Do I have to create another project and feed the outputs of DARTEl to functional and structural fields and GM/WM/CSF to the ROI tab? is that all?

My last question is about scrubbing and QC time-series. When a preprocessed dataset is being used, which output files should be used for these fields?

Thank you in advance for your help regarding any of my many questions!


Threaded View

preprocessing questions
ela Feb 24, 2021
ela Feb 26, 2021