Sparse PCA a new method for unsupervised analyses of fMRI data
Sjstrand K, Larsen R, Lund T, Madsen K
Technical University of Denmark, Copenhagen University Hospital
Exploratory analysis of functional MRI data aims at revealing known as well as unknown properties in a data-driven manner devoid of hypotheses on the time course of the hemodynamic response. Common approaches include clustering methods, principal component analysis (PCA) and in particular independent component analysis (ICA). ICA assumes that the measured activity patterns consist of linear combinations of a set of statistically independent source signals. We introduce a competing method for fMRI analysis known as sparse principal component analysis (SPCA). We argue that SPCA is less committed than ICA and demonstrate similar results, with better suppression of noise.