 |
 |
Hi, You can't directly *register* a statistic image to a structural image. You can resample a statistic image using a transform derived by registering a functional image (with grayscale intensities, looking like a brain) to a structural image. So if you are trying to do a direct registration (rather than resampling with a predetermining transformation matrix) then it will not work. You must register a functional image to the structural image, then take the matrix output from this and apply it to the statistic image. I can't really help with the initial functional data registration being "totally off" without some more information. It is completely normal for the functional and structural images to have a different number of slices. It would be very strange if this was not the case (either a very high-res functional, or a very low-res structural). So the problem must be to do with something else. Can you give some more information? If the Field Of View (FOV) is very different between the images then that can cause problems, as can distortions and signal dropouts that are common in functional images. There are different strategies to try, but we need to know more in order to help. All the best, Mark On 18 May 2011, at 16:08, Aditya Kumar Kasinadhuni wrote: > Hi all, > > I am trying to register a statistic image at the functional data resolution to a structural image (animal dataset) but am getting erroneous results. > > I ran Feat on the initial functional data and the registration step was totally off to begin with. (probably due to diff number of slices in functional and structural data may be? I am not sure if I can run Feat in a similar way as we run iot for humans) > > Can someone pelase assist me? >