That is what it means to add scanner as a nuisance variable. How unbalanced is the design?
On 06/26/2018 01:30 PM, Heidi Lindroth wrote:
Hi Bruce,
It is a cross-sectional, descriptive analysis. The data has already been collected. The participants are not evenly distributed. Would it be appropriate to regress out scanner type if I am using the raw data (extracted via DKT atlas)?
thank you,
Heidi Lindroth BSN RN PhD Candidate| School of Nursing Program Assistant | School of Medicine and Public Health, Department of Anesthesiology Predoctoral Fellow | Veterans Administration (Madison WI)
University of Wisconsin-Madison 701 Highland Avenue, Madison WI 53705 608-219-4294 | hlindroth@wisc.edu
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu *Sent:* Tuesday, June 26, 2018 12:27:35 PM *To:* Freesurfer support list *Subject:* Re: [Freesurfer] Adjusting/accounting for between scanner variance Hi Heidi what kind of study are you doing? Usually scanner is included as a nuisance regressor, but you also want to make sure that you balance your subject population across scanner as much as you can
cheers Bruce
On Tue, 26 Jun 2018, Heidi Lindroth wrote:
External Email - Use Caution
Hi there, Is there a standard method to use for adjusting for between scanner
variance? In the sample I am analyzing, 4 different
scanners were used.
Thank you,
Heidi Lindroth BSN RN PhD Candidate | School of Nursing Program Assistant | School of Medicine and Public Health, Department
of Anesthesiology
Predoctoral Fellow | Veterans Administration (Madison WI)
University of Wisconsin-Madison 701 Highland Avenue, Madison WI 53705 608-219-4294 | hlindroth@wisc.edu
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer