Hi there,
thanks for your interest in my problem here.
Actually, I really had those problems with image headers beforehand. But I
think that's solved by now. What I am doing now is the following:
1) I convert the DICOMs with SPM2 -> those ANALYZE files go to SPM2 only.
2) I do another conversion: native DICOM to ANALYZE with (X)medcon
(xmedcon.sourceforge.net) -> those ANALYZE files go to FSL.
(X)medcon shouldn't touch the headers,as far as I know .
Thus, this problem of changing global signals / scalefactors shouldn't be
an issue (I had this beforehand, that's true. You could see it in
FSLViewers "movie mode" pretty well).
I'm pretty sure I did something wrong somewhere - I just don't know where
yet. I'll re-run everything from scratch, double-checking every setting.
After that, I'll report again.
Thanks for your input,
best regards,
Cornelius Werner
On Mon, 06 Dec 2004 14:23:54 +0100, John Ashburner
<[log in to unmask]> wrote:
> After discussion with Steve S, and Mark J at FMRIB, we think that the
> most
> likely reason for the difference is the scalefactors in the image
> headers.
> If you used the DICOM conversion routine of SPM to convert Siemens Mosaic
> DICOM data, then the images will all have different scalefactors in their
> .hdr files. SPM uses these scalefactors, but FSL ignores them in data
> that
> are not in the NIFTI-1 format. Instead, the FSL format assumes that all
> the
> .img files are scaled equally. Because the images are in the SPM format,
> rather than the FSL format, you would get blobs in SPM, but not FSL.
>
> Best regards,
> -John, Steve and Mark
>
>> just out of curiosity, I analyzed a fairly complex (and long) erfMRI
>> experiment with a 3x3 factorial design with both SPM2 and FSL3.2b. I
>> tried
>> to keep everything as close to the other package as possible, i.e. I
>> adjusted both smoothing kernels to 5mm, set the high pass filter to
>> 128s,
>> used gamma function + temp. derivative in FSL (standard delay) /
>> canonical
>> hrf + tdv in SPM2, entered identical onset vectors, specified identical
>> contrasts and so on. I set the duration for my events to the actual 1.5
>> seconds in both packages, as FSL doesn't allow 0-durations.
>> Now one of the packages gave me "blobs" on a
>> corrected-for-multiple-comparisons level of 0.05, where the other didn't
>> show anything worth mentioning on a uncorrected p=0.001 level. Is that
>> possible at all, given the similarity of the statistics? Has anyone
>> experienced something similar? I am aware of the comparison done by
>> Bianciardi et al (NeuroImage 2004), but differences were marginal at
>> best...!
>
--
Cornelius Werner
Institut fuer Medizin (IME)
AG Kognitive Neurologie
Forschungszentrum Juelich
52425 Juelich
Germany
Tel. +49-(0)2461-61-8609
|