If your individual structurals don’t have a nose, it’s not surprising that SPM cannot figure out the correct nasion position for them, is it? If you coregister your individual images to the template you could take the template nose and ears as a reasonable initial guess for the fiducials and then set ‘use head points’ to ‘yes’ to get a better fit for each head. You can do that via batch so that you don’t have to type the fiducials in for each subject.
Best,
Vladimir
On 5 Jan 2014, at 10:57, David Pedrosa <[log in to unmask]> wrote:
> Hi Vladimir,
>
> thank you for your quick reply!
>
> You should just find the MNI coordinates on your individual images and use them with the ‘type’ option or the corresponding option in batch.
>
> Does it mean that I have to manually type the three fiducials for every patient? My problem with the coregistered MRI is that it cuts a little part in the front of the images after coregistration. It doesn't affect the brain but I cannot provide fiducials e.g. for the nasion. What am I doing wrong?
> Regards
>
> David
>
> Am 03.01.2014 18:16, schrieb Vladimir Litvak:
>> Dear David,
>>
>> It sometimes happens that the standard fiducially are not transformed to the individual space properly. You should just find the MNI coordinates on your individual images and use them with the ‘type’ option or the corresponding option in batch. This is more precise in any case. I would also recommend to coregister and reslice your individual images to the template before using them for head modelling.
>>
>> Best,
>>
>> Vladimir
>>
>> On 3 Jan 2014, at 14:55, David Pedrosa <[log in to unmask]> wrote:
>>
>>> Hi all,
>>>
>>> I'm currently working on EEG data to analyse evoked resposes for different motor paradigms. Right now, I'm about to coregister EEG and individual MRI data. I already managed to do so for the majority of subjects. However, I'm stuck with some subjects for which the detection of fiducials fails (it only appears NaN). Besides one of the subjects shows one fiducial which is ridiculously far away from the MRI. I read in this list to try to coregister with the offered template, but this did not work (I tried both versions spm8 and spm12b). Is there something else I could do? Or does anybody know where the problem might be? Thanky in advance
>>>
>>> Best wishes
>>>
>>> David
>
|