On Clustering fMRI Time Series Using Potts and Mixture Regression Models

• Activity Number: 318
• Type: Contributed
• Date/Time: Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Jing Xia*+ and Feng Liang and Yongmei (Michelle) Wang
• Companies: University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign
• Keywords: Potts model ; Restoration Maximization Algorithm ; Mixture Models ; Clustering ; Functional Connectivity

Abstract

We propose a model-based clustering method for functional magnetic resonance imaging (fMRI) data analysis. To incorporate spatial information, Potts model is introduced to represent spatial interactions of neighboring voxels, and to integrate the temporal regression modeling into one single unified model. The estimation of the parameters is achieved through restoration maximization (RM) algorithm for computation efficiency and accuracy. Additional features of our method include: the optimal number of clusters can be automatically determined from AIC/BIC; the global trends and the informative paradigms of the data are extracted by a dimension reduction algorithm based on principal component analysis (PCA) and a statistical significance test. Simulated data and real applications demonstrate that our approach can lead to robust and sensitive detection of functional clusters and networks.