 |
 |
Hello Tony, I suppose the coregistration step didn't work properly, therefore the EPI seems to be shifted upwards. Check whether this shift already exists between coregistered mean and T1 image. If yes, then you should reorient the structural image using the "Display" function manually (after automatic coregistration). In my experience, this happens rather often (even if subjects did not move between acquisition of T1 volume and EPI series), but so far, I only had shifts along the dorsal - ventral axis of about 0 - 4 mm (so more or less corresponding to the slice thickness). Everything else was ok. Therefore, it is rather easy to correct: - load mean and T1 with "Check reg" - look for some anatomical landmark which is clearly visible on both the two volumes, e.g. the upper border of the corpus callosum, so that you can estimate the misplacement quite easily - then load the T1 volume with "Display", add the estimated distance, click on "reorient" and select this T1 volume again (otherwise nothing is changed) Don't forget to check whether the mean and T1 volume are in accordance afterwards, and especially whether you moved the T1 volume into the correct direction (in case it was the wrong direction, load the volume with "Display" again and enter a number twice as large and with opposite sign). Therefore it's never a bad idea to make a copy of the T1 volume before you reorient it. Best, Gabor