 |
 |
Hi - if this appears to have worked then it probably has! However it's possible that your model-based confound removal *might* not have modelled the confound quite as well as ICA would have - so you could try that as well - i.e. use MELODIC and then fsl_regfilt. Cheers. On 29 Dec 2010, at 09:19, Jasper Looijestijn wrote: > Dear Steve, > > Earlier I posted this message to the list, but did not get a response. Can you inform me on this issue? > > We are attempting to remove a 0.41 Hz noise artifact from our 4D BOLD fMRI data. The model that describes the artifact is: > > EV1: > 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0… > EV2: > 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0….. > EV3: > 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0….. > > We used the above design for the <confound.mat> used in the commandline script “unconfound” and applied the script to all our data. Is this a correct approach to remove the noise-artifact? > On inspection of our unconfounded data we seem to have succeeded to a large extent. Does anyone have experience with the effects of this script on the data? Is there a downside to this approach we should be aware of? > > With kind regards, > > Jasper Looijestijn. > > --------------------------------------------------------------------------- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ---------------------------------------------------------------------------