Dear Tae,
There's the spatial resolution to be taken into account, but that's not all.
To give you an example from PET where spatial resolution is typically some ~5mm these days, with 5HT1A tracers you can see the dorsal raphe nucleus very well, whereas you would be hard pressed to define it on MRI with a resolution of ~1mm. The reason is the very high contrast-to-noise ratio for such selective tracers - there's lots of 5HT1A binding in the dorsal raphe and next to nothing around it, so you will see tiny structures even when they are smaller than your system's nominal resolution (the signal will just be averaged over several neighbouring voxels).
I can't comment on your particular study because I don't know the tracer you're using (and no, I'm not your reviewer ;-) ) but it might be an entirely sensible suggestion to look at a hypthalamic signal, or at other small structures.
In addition to these technical system issues, do not forget you are also typically averaging over lots of subjects, and if spatial normalisation is good, with good resulting overlap of small stuctures, you may be able to see differences between patients and controls in structures smaller than your system's nominal resolution.
Good luck,
Alexander
-------------------------------
Dr Alexander Hammers, MD PhD
MRC Clinician Scientist Fellow
Honorary Clinical Lecturer
Clinical Sciences Centre
Room 243, MRC Cyclotron Building
Faculty of Medicine, Imperial College London
Hammersmith Hospital, DuCane Road, London W12 0NN
and
Honorary Lecturer in Neurology
Department of Clinical and Experimental Epilepsy
National Hospital for Neurology and Neurosurgery/ Institute of Neurology,
UCL
33 Queen Square, London WC1N 3BG
Telephone +44-(0)20-8383-3162 (ext./direct line -3704 or -2651)
Fax +44-(0)20-8383-1783 /-2029
Email [log in to unmask], [log in to unmask]
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]On Behalf Of [log in to unmask]
Sent: 21 February 2005 07:16
To: [log in to unmask]
Subject: [SPM] intrinsic spatial resolution of SPECT scanner and extent thresholding
Dear experts of neuroimaging.
I'd like to ask a question that how I determine the extent threshold of SPM result.
According to the height threshold, the significant cluster size would vary. So, I
thought ignoring the smaller clusters than intrinsic spatial resolution of SPECT scanner (6.9 mm)
is irrational. But, some reviewers for my manuscript disappointed me with this problem (pointing
out the small size of brain structures(i.e. hypothalamus...)
--------------------------------------------------------------------------------------------------------------------
Woo Suk, Tae
Neuroimaging Lab. Dept. of Neurology, Samsung Medical Center, Seoul, Korea
Ph.D. Candidate, Computational Neuroimage Analysis Lab. Dept. of Biomedical
Engineering, Hanyang University
Tel: 82-2-3410-2738
Fax: 82-2-3410-2759
email: [log in to unmask]
--------------------------------------------------------------------------------------------------------------------
|