Comparison of statistical parametric mapping method and scaled subprofile model for functional neuroimage analysis*

Two common analytic techniques for functional neuroimages are Statistical parametric mapping (SPM) and Scaled subprofile model based on principle component analysis (SSM/PCA). SPM is based on mass univariate testing in which signals from each region are compared between two groups. SSM/PCA is a multivariate PCA-based algorithm that construct the disease-related pattern that discriminate between patients and controls.
We developed analytical relation between the SPM and SSM/PCA results. We tested this relation on the results of SPM and SSM/PCA, performed on simulated images with various imaging scenarios addressed (various number of images, noise levels, image patterns) and on clinical 18F-flourodeoxyglucose positron emission tomography images of Alzheimer's dementia patients.
Investigation of SPM and SSM/PCA methodology revealed that SSM/PCA and SPM results are related: SSM/PCA disease-related pattern is proportional to the logarithm of SPM's t-map, increased by 1 and corrected by a factor depending on the number of images and imaging noise level. Tests with simulated and clinical images verified this relation.
This work confirmed that the results of SPM and SSM/PCA on the same data are not independent, despite the fundamental differences between those two methods.