SDM   Seed-based d Mapping
formerly "Signed Differential Mapping"
Share on Facebook:   Facebook

SDM calculations

Linear model analyses

Linear models might be useful to several aims such as comparing two or more groups, controlling for potential confoundings, or assessing the heterogeneity of the findings by means of meta-regressions.

To calculate a linear model you need to specify the model (up to four variables to use) and the hypothesis. You can also specify a filter for subgroup analyses.

To calculate a linear model

Press the button [Linead Model]


Select [Linear Model] in the Statistics menu, to open the following dialog:

Linear model dialog


The MLE step of the linear model calculation, which can take long in large meta-analyses, can be run in parallel using concurrent threads. Note: is not advisable to use more concurrent threads than the available in your pc. You can learn the number of available threads in your pc at the 'multithreading' tab of the preferences windows.

Command-line and batch usage

mi_lm model,hypothesis,number_of_imputations,filter where the hypothesis is the list of values to use separated by the character + and the model is the list of covariates to use separated by the character +.


model_name = mi_lm YBOCS,0+1+0+0+0,50,adults

Note: In order to modify the number of concurrent threads while using the command-line, you need to modify by hand the value of nthreads in the file sdmpsi_params.xml, that is created after the preprocessin


(original algorithms): Radua J and Mataix-Cols D. Voxel-wise meta-analysis of grey matter changes in obsessive-compulsive disorder. Br J Psychiatry 2009; 195:393-402. link .

(comparisons and linear models): Radua J, van den Heuvel OA, Surguladze S and Mataix-Cols D. Meta-analytical comparison of voxel-based morphometry studies in obsessive compulsive disorder vs other anxiety disorders. Arch Gen Psychiatry 2010; 67:701-711. link .

(effect-sizes and variances): Radua J, Mataix-Cols D, Phillips ML, El-Hage W, Kronhaus DM, Cardoner N and Surguladze S. A new meta-analytic method for neuroimaging studies that combines reported peak coordinates and statistical parametric maps. Eur Psychiatry 2012; 27:605–611. link .

(current method): Albajes-Eizagirre A, Solanes A, Vieta E and Radua J. Voxel-based meta-analysis via permutation of subject images (PSI): Theory and implementation for SDM. NeuroImage 2019; 186:174. link