There are much more sensible ways to increase SNR than smoothing in the
volume. If you must smooth, you need to ensure that you are smoothing
within the same structure. 3D smoothing in the volume smooths across
white matter, CSF, and adjacent greymatter (which may or may not be a part
of the same cortical area). If you move to smoothing in 2D on the
surface, that avoids some of these problems, but still will have you
smoothing across cortical areal boundaries (or subcortical sub-nuclei if
you are talking about smoothing in the volume that is constrained to
subcortical structures). A much better way to increase SNR is just to
average within the brain areas. We have written about these issues
extensively: https://www.nature.com/neuro/journal/v19/n9/abs/nn.4361.html,
https://www.pnas.org/content/115/27/E6356.short.
Gaussian random field theory has a number of issues that were explored in
this paper: https://www.pnas.org/content/113/28/7900. Permutation-based
methods are more statistically valid and donıt require smoothing.
Iım not sure what you mean by a higher level linear trend. If you mean
higher frequencies, yet sICA+FIX can remove these if they appear together
with other artifactual signals in spatially specific locations.
Matt.
On 2/27/19, 6:57 AM, "FSL - FMRIB's Software Library on behalf of Sam W."
<[log in to unmask] on behalf of [log in to unmask]> wrote:
>Thank you Matthew that is very helpful.
>However isn't the reason for smoothing to increase SNR, which is why it's
>so frequently applied? Smoothing the data is also required if ones plans
>to use Random Field Theory. Or do you use other corrections that don't
>depend on RFT?
>Also applying linear detrending will only eliminate linear trends. What
>if there are other higher level trends? I think the high pass filter
>implemented in SPM, which uses a discrete cosine basis set, is able to
>account for these, but I might be wrong. Do you think these confounds can
>be removed with sICA+FIX?
>
>Best regards,
>Sam
>
>########################################################################
>
>To unsubscribe from the FSL list, click the following link:
>https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
________________________________
The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|