Hi again Negar,
You may not have enough power to detect small differences with only 8 in your normal control group. You may already know what I am saying but here are a few
of my suggestions:
1. How much smoothing did you apply to the scans? Try increasing the smoothing to increase the signal to noise ratio (maybe 16-24mm). This will reduce the regional
specificity of any results though.
2. Check your method with a SPECT scan that has obvious abnormalities to get an idea how sensitive the SPM might be.
3. Check that you used proportional normalization and relative threshold masking (about 0.8 should be ok).
4. Try using really liberal statistical thresholds say 0.05 uncorrected with no cluster extent threshold, to see if there might be any subtle signal anywhere
that would correlate to the patient or the qualitative reading of the scans.
Good luck,
Kevin London
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]
On Behalf Of Negar Chabi
Sent: Monday, 12 January 2015 9:36 PM
To: [log in to unmask]
Subject: [SPM] SPECT individual analysis
Dear all
I use SPM to analyze SPECT images and find abnormalities individually, by comparing each patient SPECT image with the normal control group images (To start we have just 8 healthy control images).
For preprocessing I perform Coregistration, Normalization and Smoothing.
After that for model specification I selected two sample t-test. And then selecting all the healthy control group as "Group 1 scans" and one of the patients for “Group 2 scans”. In addition I select [-1 1] , [1
-1] for Hyper perfusion and Hypo perfusion respectively. But , finally I receive "no supra threshold clusters" for all of the patients after comparing by normal group individually. I don't know why this happens, and how can I solve it?
The final results were attached.
I greatly appreciate any advice.
Best
--
Negar Chabi
Msc of medical engineering
Persian Gulf Nuclear Medicine Research Center
Bushehr University of Medical sciences
Views expressed in this message and any attachments are those of the individual sender, and are not necessarily the views of The Sydney Children's Hospitals Network
This note also confirms that this email message has been virus scanned and although no computer viruses were detected, The Sydney Children's Hospitals Network accepts no liability for any consequential damage resulting from email containing computer viruses.