Abstract:
|
Independent component analysis (ICA) is an unsupervised learning method popular in functional magnetic resonance imaging. Group ICA has been used to identify biomarkers in neurological disorders including autism spectrum disorder and dementia. However, current methods use a PCA step that may remove low-variance features. Linear non-Gaussian component analysis (LNGCA) enables dimension reduction and feature estimation including low-variance features simultaneously in single-subject fMRI. We present a group LNGCA model to extract group components shared by more than one subject. To determine the total number of components, we propose a parametric bootstrap test that samples spatially correlated Gaussian components to match the spatial dependence observed in data. In simulations, our estimated group components achieve higher accuracy compared to group ICA. We apply our method to a resting-state fMRI study on autism spectrum disorder for 342 children (252 typically developing, 90 with high functional autism), where the group signals include resting-state networks. This novel approach to matrix decomposition is a promising direction for feature detection in neuroimaging.
|