BrainVoyager Discussion Forum
  Getting started with projects, protocols and data analysis
  Spatial domain vs. frequency domain.

Post New Topic  Post A Reply
profile | register | preferences | faq | search

UBBFriend: Email This Page to Someone! next newest topic | next oldest topic
Author Topic:   Spatial domain vs. frequency domain.
aristodos
Member
posted 22 December 2003 20:01     Click Here to See the Profile for aristodos   Click Here to Email aristodos     Edit/Delete Message   Reply w/Quote
Hi,
Just one question... or two. I am doing spatial and temporal smoothing in 3D. However, what´s the difference between spatial and frequency domain on spatial smoothing? Any reference to understand this difference and to decide which apply? I am applying spatial smoothing in frequency domain. However, I understand much better the application in spatial domain. I have a conflict, help!
And is there any difference if I apply smoothing (spatial) before or after normalizing? I am doing it after normalization but... just a question... why then?
And the last question, I read once that there isn´t any difference between applying spatial smoothing in 2D or 3D. However, I wonder if is there a difference in the size of the HWFM that I would use when I do smoothing in 2D or 3D? It looks like it doesn´t make any difference but the most I read, the most questions I have.

Thanking you in advance.

Best,

Henk Jansma
Member
posted 12 January 2004 12:08     Click Here to See the Profile for Henk Jansma   Click Here to Email Henk Jansma     Edit/Delete Message   Reply w/Quote
Dear Aristodos,

The difference between spatial smoothing in the spatial domain and in the frequency domain, is that with smoothing in the frequency domain, the data is fourier transformed. The results of both ways of smoothing are more or less the same. When smoothing with a small kernel, I would like to advise to use the spatial domain, instead of the frequency domain, since the fourier transformation takes some time. For detailed information, please have a look at the BrainVoyager 2000 User's Guide, under section: "FMR Preprocessing - Update".

We would like to advise to spatially smooth (if at all) on the non-Talairach data. Warping the cortex into Talairach space, has the side effect of 'blurring' the data. If you have smoothed your VTC data only slightly, and you found the data isn't smoothed enough, you can do the smoothing on the VTC data set. This saves you the trouble of creating a new VTC (with different smoothing parameters). The smoothing is done (also on the FMR) in 3D. BrainVoyager makes a 'stack' of the FMR slices, and performs the preprocessing steps. For detailed information, please have a look at the BrainVoyager 2000 User's Guide, under section: "FMR Preprocessing - Update".

With kind regards,

------------------
Henk Jansma

Brain Innovation
Universiteitssingel 40
6201 BC Maastricht
The Netherlands
+31 43 388 40 64 (phone)
+31 43 388 41 25 (fax)
+31 6 125 70 144 (mobile)

You can reach me by e-mail, snail-mail or phone.

All times are ET (US)

next newest topic | next oldest topic

Administrative Options: Close Topic | Archive/Move | Delete Topic
Post New Topic  Post A Reply
Hop to:

Contact Us | BrainVoyager.com

Copyright © Rainer Goebel 2001 - 2006

Powered by Infopop www.infopop.com © 2000
Ultimate Bulletin Board 5.47d