 |
 |
> I am trying to register the subject's brain extracted structural images to > the standard single subject T1 template provided by SPM99. This is very unlikely to work. The mean squared difference cost function does not work well at all for registering a skull-stripped brain with a template image that has the scalp included. > However the > problem I am facing is that I am unable to check whether the subject > structural image is in the same orientation as the standard template. I > have tried reorienting the subject's image using the Display option in > SPM99 to get it in the same orientation as the standard template. However I > am unsure about the right/left side of the images. When I open the image in > MRIcro; they seem to be flipped as compared to the image seen after > reorienting using the Display option. 1. How do I get the subject's image > in the same orientation as the standard image. Try the Check Reg button. If this seems to show the images in register, and the handedness of your images matches the defaults.analyze.flip variable (1 for left-handed, and 0 for right-handed), then all should be well. On second thoughts - as you are using a very old version of SPM (SPM99), you need to ensure that the orientation of the spatially normalised images is correct. This will involve ensuring that the value of sptl_Ornt in the defaults is set appropriately. These provide starting estimates for the affine registration part of the spatial normalisation. In SPM99 (unlike more recent versions), images were sometimes flipped during spatial normalisation, in order that the right-side of a spatially normalised brain appears on the right side of the screen when you display the image. SPM can't magically figure out the handedness of image data, so it is important to get this right before you begin. If the DICOM converter in SPM is used, then all should be correct, as this looks at the orientation information in the image headers. > > For your reference I am attaching screen shots of the images using MRIcro > and Display option in SPM99. I haven't used MRIcro for a long time now, so I can't comment. Looking at the images from SPM99, it appears that your voxel sizes in the x direction may not be quite correct. These need to be set properly. As you are a new user, I would urge you to use a more recent version of SPM. Best regards, -John