 |
 |
> My second question concerns the use of suitable mask after applying > normalize to MNI function for fMRI data to get ride of potential aliasing > effects.Is it better to use normalized template of dartel as an > explicit mask (while doing second level analysis) or to apply the > deformation to the first level subject-specific mask and then multiply with > the smoothed normalized images created by normalize to MNI function? > These are not really aliasing effects. It's the effect of missing data in the images, where SPM tries to guess about what values should be there. Usually, these regions contain zeros, which cause even greater artifacts after smoothing, as some subjects' scans will have data, whereas others have zeros. Because this seems to be a concern for so many, I'll have a go at including a fix for the data so that those voxels with very little signal are set to zero. Best regards, -John --00163630f77d500ad5049c7a214f Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable <br><div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"m= argin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><table cells= pacing=3D"0" cellpadding=3D"0" border=3D"0"><tbody><tr><td valign=3D"top" s= tyle=3D"font:inherit"> <div>I am wondering how to use spm_get_space to affine normalize the=A0DART= EL template in MNI space? Shall one use *Template_6 as P and *_2mni.mat as = M?</div> <div>Isn't it possible to use normalize_write function instead?</div></= td></tr></tbody></table></blockquote><div><br></div><div>You could write th= e matrix in *_2mni.mat into the header of the Template_6.nii image. =A0The = spm_get_space function would do that.</div> <div><br></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex= ;border-left:1px #ccc solid;padding-left:1ex;"><table cellspacing=3D"0" cel= lpadding=3D"0" border=3D"0"><tbody><tr><td valign=3D"top" style=3D"font:inh= erit"> <div>=A0</div> <div>My second question concerns the use of suitable mask=A0after applying = normalize to MNI function for fMRI data to get ride of potential=A0aliasing= effects.Is it better to use normalized template of dartel as an explicit= =A0mask (while doing second level analysis) or to apply the deformation to = the=A0first level subject-specific mask and then multiply with the smoothed= normalized images created=A0 by normalize to=A0MNI function?=A0</div> </td></tr></tbody></table></blockquote><div><br></div><div>These are not re= ally aliasing effects. =A0It's the effect of missing data in the images= , where =A0SPM tries to guess about what values should be there. Usually, t= hese regions contain zeros, which cause even greater artifacts after smooth= ing, as some subjects' scans will have data, whereas others have zeros.= </div> <div><br></div><div>Because this seems to be a concern for so many, I'l= l have a go at including a fix for the data so that those voxels with very = little signal are set to zero.</div><div><br></div><div>Best regards,</div> <div>-John</div></div> --00163630f77d500ad5049c7a214f--