Search for Default Network Using Likelihood-Based Population Independent Component Analysis

• Activity Number: 116
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Imaging
• Author(s): Lei Huang*+ and Shaojie Chen and Huitong Qiu and Ani Eloyan and Ciprian M. Crainiceanu and Brian Caffo
• Companies: Johns Hopkins University and The Johns Hopkins University and The Johns Hopkins University and Johns Hopkins Bloomberg School of Public Health and The Johns Hopkins University and Johns Hopkins University
• Keywords: independent component analysis ; parallel computing ; default network ; resting state fMRI

Abstract

Independent component analysis (ICA) is a computational method for separating a multivariate signal into additive subcomponents supposing the mutual statistical independence of source signals. The technique is used heavily in several scientific research areas including acoustics, electrophysiology and functional neuroimaging. We propose a scalable two-stage iterative true group ICA methodology for analyzing population level fMRI data. The method is based on likelihood estimators of the underlying source densities. It can be applied to a large group of subjects since the memory requirements are not restrictive. We also use the parallel computing techniques in the estimation algorithm to improve the time efficiency. The proposed method is applied to a large collection of resting state fMRI datasets. The results show that the default brain networks are recovered by the proposed algorithm.