 |
 |
Ok! Thanks again, best, José El vie., 7 sept. 2018 a las 16:24, Janine Bijsterbosch (< [log in to unmask]>) escribió: > Hi, > > As far as I know, the salience network is part of most templates. You > would typically want to include *all* networks in the dual regression > though, because the multiple regression takes into account the other > networks when estimating the subject-specific maps for the salience > network. So even if you’re only interested in the salience network, I would > recommend including all maps in the dual regression. You can either switch > off permutation testing (stage 3) and manually run it only for the salience > network, or you can delete stage 3 results for all other networks post hoc. > > Cheers, > > Janine > > > On 7 Sep 2018, at 15:11, Jose Pineda <[log in to unmask]> wrote: > > Thank you Janine, > > Are you aware of an available template that could be used to study the > salience network as a single IC? > > Thanks! > José > > > El vie., 7 sept. 2018 a las 8:36, Janine Bijsterbosch (< > [log in to unmask]>) escribió: > >> Hi, >> >> You should definitely *not* create group IC maps based on the healthy >> controls only. This will introduce a statistical bias into your analysis >> because your maps are driven by just one of your subject groups. >> >> The best option is probably to include all 45 scans and obtain group IC >> maps from this. Taking this approach has the advantage of finding >> study-specific noise maps, and including these in subsequent dual >> regression analysis provides second-level cleanup. It is also a good option >> to use templates from existing datasets. It isn’t really the case that one >> set of templates is more suitable than another, so you can pick for example >> based on how well the templates match the existing literature in your field. >> >> Best wishes, >> >> Janine >> >> >> On 6 Sep 2018, at 18:10, Jose Pineda <[log in to unmask]> wrote: >> >> Hello, >> >> I'm running a dual regression analysis on a local dataset with Ncontrol >> 15, Npatients 15, being the patients scanned under two conditions. >> Regarding the creation of IC group networks I see that there are two >> options, either to include all scans (N45), or just the healthy controls >> group, correct?. >> Could I also use network templates from other datasets as the one >> published in Smith et al. PNAS 2009? If yes, which templates will be the >> more suitable for this matter and what options do I have? >> >> Thanks >> José >> >> ######################################################################## >> >> To unsubscribe from the FSL list, click the following link: >> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 >> >> >> ----- >> Dr Janine Bijsterbosch >> Postdoctoral Researcher >> FMRIB Centre, University of Oxford >> John Radcliffe Hospital >> Oxford, United Kingdom >> [log in to unmask] >> >> >> ------------------------------ >> >> To unsubscribe from the FSL list, click the following link: >> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 >> > > > -- > José Angel Pineda-Pardo > Postdoctoral Researcher > hmCINAC - HM Puerta del Sur > Av Carlos V 70, Móstoles, Spain > > ------------------------------ > > To unsubscribe from the FSL list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 > > > ----- > Dr Janine Bijsterbosch > Postdoctoral Researcher > FMRIB Centre, University of Oxford > John Radcliffe Hospital > Oxford, United Kingdom > [log in to unmask] > > > ------------------------------ > > To unsubscribe from the FSL list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 > -- José Angel Pineda-Pardo Postdoctoral Researcher hmCINAC - HM Puerta del Sur Av Carlos V 70, Móstoles, Spain ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1