Dear experts,

I have run two independent meta-analyses with SDM 5.141 and all seems to work fine (which means that I do not get any error messages). However, there are two aspects which I would like to clarify.

1.) When looking at the heterogeneity maps I always get one large cluster with more or less the whole brain. This is rather troubling and might suggest that

• either the results of my studies are that heterogeneous (which might in fact be true), or
• that something is off with my calculations.

The execution log lists the  following command:
About to execute: s_exec: "C:\SdmGui-win64\bin\windows64/sdm_core" threshold MyMean_WithinGroups_z_QH,p,0.005000,1.000000,10,0

Please find attached the results of these calculations (slide 1).

2.) The meta-analytic results seem fine when only considering the tables. I did use a threshold of k>50 voxels. However, looking at the images in MRIcroN I get several tiny clusters (k might be 1 voxel?), which are not listed in the tables. Is there an explanation for this divergence of the tables and the maps?

The execution log lists the following command:
About to execute: s_exec: "C:\SdmGui-win64\bin\windows64/sdm_core" threshold MyMean_WithinGroups_z,p,0.005000,1.000000,50,0

Please find attached a screenshot of the table and map (slide 2).

Thank you very much in advance and all the best,
Lars

Dear experts,
 
upon further examination of our data, we realize that there is one study with abnormal mean estimates and variances at all voxels of the brain (e.g., mean 818.26165; variance 14482.051758; the rest of studies ranges between -0.4 and 0.5). For this study we included an original statistical parametric map (SPM) in our analysis. The map itself looks absolutely fine.
 
How would you suggest to proceed? Should we exclude this study?  
 
Thanks, Lars
 
 

Dear Lars,

I'm sorry to hear that. Please send me the file by email and I'll have a look...

With the best wishes,

Joaquim