Mingwu Jin1, Rajesh Nandy2, Dietmar Cordes1
1Radiology, University of Colorado Denver, Aurora, CO, USA; 2Biostatistics and Psychology, UCLA, Los Angeles, CA, USA
Local canonical correlation analysis (CCA) is a multivariate method that has been proposed to determine more accurately activation patterns in fMRI data. One drawback of CCA is that, unlike the general linear model (GLM), an arbitrary linear contrast of the temporal regressors has not been incorporated in the CCA formalisms. In this research we show how to extent CCA so that an arbitrary linear contrast of the temporal regressors can be computed similar to a t-statistic in GLM.