Overview of SPM 2 From buttons to code

Overview of SPM 2 From buttons to code Eamonn Walsh & Domenica Bueti

SPM Statistical Parametric Mapping • SPM examines every voxel location across all images and computes a parametric map • Data reduction – condensing information from a number of individuals in a statistically meaningful way

Data transformations Image time-series Kernel Design matrix Realignment Smoothing General linear model Statistical parametric map (SPM) Statistical inference Normalisation Gaussian field theory p <0. 05 Template Parameter estimates

Spatial Preprocessing Model Specification and Parameter Estimation

Spatial Preprocessing

Spatial Preprocessing - Realign • Aligning hundreds of 3 D brain volumes per single subject • To create a ‘mean image’

Spatial Preprocessing - Coregister • Matching the functional scan to the structural scan for the same individual

Spatial Preprocessing – Slice Timing • Take first or middle slice as reference slice • Shift all other slices in time to match this reference slice

Spatial Preprocessing - Normalise • The same voxels in different brains are aligned to represent the same anatomical location. • Template

Spatial Preprocessing - Smooth • Blur the images prior to statistical analysis • FWHM Full width half maximum

Spatial Preprocessing - Segment • Images are segmented into grey and white matter maps

Image File fm 00223_004. img Realigned rfm 00223_004. img Realigned, Coregistered, , rrfm 00223_004. img Realigned, Coregistered, Normalised, wrrfm 00223_004. img Realigned, Coregistered, Normalised, Smoothed swrrfm 00223_004. img

Spatial Preprocessing

SPM Demo Spatial Preprocessing Guided Tour