Old message but it look like you never got a reply.
I think the FIR approach is likely your best bet. It should do a decent job of estimating the response at each TR without getting into over modelling.
Here is an older example from some former colleagues of mine who used a different approach:
https://www.ncbi.nlm.nih.gov/pubmed/14568485
In my past life we also tried estimating some patient specific HRF responses using a fourier basis set. This allows for total flexibility in the HRF shape (and we only applied it to regions where we knew we had an response with the canonical HRF), but we found that it could be difficult to estimate a reliable clean response on an individual basis.
Not sure about FLOBS, never tried it, but it looks to me like it is focused on using a pre-specified HRF shape rather than calculating a specific response?
Best of luck,
Colin Hawco, PhD
Neuranalysis Consulting
Neuroimaging analysis and consultation
www.neuranalysis.com
[log in to unmask]
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Stephen Bailey
Sent: February-09-17 9:29 AM
To: [log in to unmask]
Subject: [FSL] Estimating Width, Time to Peak for HRF
Hello all --
I have some event-related data, and I am interested in getting estimates of different aspects of the HRF. Instead of a parameter estimate for response amplitude only, I would like to get estimates for Width and Time-To-Peak as well (as discussed, e.g. in Lindquist et al., 2009 in NeuroImage).
One way of doing this would be to build an FIR model then measure these properties based on the estimated timecourse -- would this be the best approach implementable in FSL? I wasn't sure if there was some way to calculate these measures based on FLOBS basis functions.
Any thoughts would be greatly appreciated!
Stephen Bailey
Vanderbilt University
|