Week 3/6

[02agarwalt]

@02agarwalt

• [x] example paper/method using epi-epi registration
• [x] find a MNI152 space EPI template.
• [x] rerun entire fngs pipeline on minimum 1 dataset as is, compute discriminability
• [x] rerun entire fngs pipeline on minimum 1 dataset with epi_reg replacing the fmri to t1w registration step, compute discriminability
• [x] present a single plot which shows these discriminability scores side by side so that we can make an informed decision.

Original paper Deliverable

@ebridge2

• [x] someething about specs specs file gists

@ewalke31

• [x] For task-based fMRI datasets, verify that components from white matter using our current compcor method do not correlate with the task-based stimulus (for approx. 10 subjects, 3-4 tasks each). Deliverable: R Markdown/Python Notebook describing experiments & results w/ plots. By Thursday: Try doing this for at least one of the subjects. Thurs: https://neurodatadesign.github.io/fngs/ewalke31/nuisance_correction/week_0306/pcasvd.html Mon: https://neurodatadesign.github.io/fngs/ewalke31/nuisance_correction/week_0306/nuiscorr_wm.html