Local canonical correlation analysis (CCA) is a multivariate method that has been proposed to more accurately determine activation patterns in fMRI data. In its conventional formulation, CCA has several drawbacks that limit its usefulness in fMRI. A major drawback is that, unlike the general linear model (GLM), a test of general linear contrasts of the temporal regressors has not been incorporated into the CCA formalism. To overcome this drawback, a novel directional test statistic was derived using the equivalence of multivariate multiple regression (MVMR) and CCA. This extension will allow CCA to be used for inference of general linear contrasts in more complicated fMRI designs without reparameterization of the design matrix and without reestimating the CCA solutions for each particular contrast of interest. With the proper constraints on the spatial coefficients of CCA, this test statistic can yield a more powerful test on the inference of evoked brain regional activations from noisy fMRI data than the conventional t-test in the GLM. The quantitative results from simulated and pseudoreal data and activation maps from fMRI data were used to demonstrate the advantage of this novel test statistic. 1. Introduction The General Linear Model (GLM) is a widely used mass univariate analysis method to determine brain activations in functional magnetic resonance imaging (fMRI) because of its simplicity in both estimation and inference and its greater sensitivity to regional effects than global multivariate analyses [1]. The least-squares (LS) solution of the GLM is the minimum variance unbiased (MVU) estimator when Gaussian white noise assumption is satisfied, otherwise the weighted LS solution (using the inverse of the noise covariance matrix) becomes the best linear unbiased estimator (BLUE) [2]. The estimated parameters and their variances are used to construct various contrast statistics, either or , to test the null hypothesis of effects of interest. Another popular approach to analyze fMRI time series uses the correlation coefficient [3]. The statistical significance of the correlation coefficient is equivalent to a -statistic testing for a regression on one single regressor [4]. The correlation coefficient is more restricted in assessing the significance of regional effects than the -test in fMRI data analysis because the correlation coefficient does not allow more than one regressor to be included for a direct calculation. It is known, however, that the partial correlation coefficient is also equivalent to a -test and thus could potentially be used
References
[1]
K. J. Worsley and K. J. Friston, “Analysis of fMRI time-series revisited—again,” NeuroImage, vol. 2, no. 3, pp. 173–181, 1995.
[2]
S. M. Kay, Fundamentals of Statistical Signal Processing: Estimation Theory, PTR Prentice Hall, Englewood Cliffs, NJ, USA, 1993.
[3]
P. A. Bandettini, A. Jesmanowicz, E. C. Wong, and J. S. Hyde, “Processing strategies for time-course data sets in functional MRI of the human brain,” Magnetic Resonance in Medicine, vol. 30, no. 2, pp. 161–173, 1993.
[4]
B. A. Ardekani and I. Kanno, “Statistical methods for detecting activated regions in functional MRI of the brain,” Magnetic Resonance Imaging, vol. 16, no. 10, pp. 1217–1225, 1998.
[5]
K. J. Worsley, A. C. Evans, S. Marrett, and P. Neelin, “A three-dimensional statistical analysis for CBF activation studies in human brain,” Journal of Cerebral Blood Flow and Metabolism, vol. 12, no. 6, pp. 900–918, 1992.
[6]
K. J. Worsley, S. Marrett, P. Neelin, A. C. Vandal, K. J. Friston, and A. C. Evans, “A unified statistical approach for determining significant signals in images of cerebral activation,” Human Brain Mapping, vol. 4, no. 1, pp. 58–73, 1996.
[7]
T. Q. Duong, E. Yacoub, G. Adriany et al., “High-resolution, spin-echo BOLD, and CBF fMRI at 4 and 7？T,” Magnetic Resonance in Medicine, vol. 48, no. 4, pp. 589–593, 2002.
[8]
O. Friman, J. Cedefamn, P. Lundberg, M. Borga, and H. Knutsson, “Detection of neural activity in functional MRI using canonical correlation analysis,” Magnetic Resonance in Medicine, vol. 45, no. 2, pp. 323–330, 2001.
[9]
O. Friman, M. Borga, P. Lundberg, and H. Knutsson, “Detection of neural activity in fMRI using maximum correlation modeling,” NeuroImage, vol. 15, no. 2, pp. 386–395, 2002.
[10]
O. Friman, M. Borga, P. Lundberg, and H. Knutsson, “Adaptive analysis of fMRI data,” NeuroImage, vol. 19, no. 3, pp. 837–845, 2003.
[11]
R. Nandy and D. Cordes, “Improving the spatial specificity of canonical correlation analysis in fMRI,” Magnetic Resonance in Medicine, vol. 52, no. 4, pp. 947–952, 2004.
[12]
M. Jin, R. Nandy, and D. Cordes, “Two local constrained canonical correlation analysis methods for fMRI,” in Proceedings of the ISMRM 17th Scientific Meeting, p. 1708, 2009.
[13]
M. Ragnehed, M. Engstr？m, H. Knutsson, B. S？derfeldt, and P. Lundberg, “Restricted canonical correlation analysis in functional MRI-validation and a novel thresholding technique,” Journal of Magnetic Resonance Imaging, vol. 29, no. 1, pp. 146–154, 2009.
[14]
S. Das and P. K. Sen, “Restricted canonical correlations,” Linear Algebra and Its Applications, vol. 210, no. C, pp. 29–47, 1994.
[15]
S. Smith, M. Jenkinson, C. Beckmann, K. Miller, and M. Woolrich, “Meaningful design and contrast estimability in FMRI,” NeuroImage, vol. 34, no. 1, pp. 127–136, 2007.
[16]
C. E. Metz, “Basic principles of ROC analysis,” Seminars in Nuclear Medicine, vol. 8, no. 4, pp. 283–298, 1978.
[17]
J. A. Sorenson and X. Wang, “ROC methods for evaluation of fMRI techniques,” Magnetic Resonance in Medicine, vol. 36, no. 5, pp. 737–744, 1996.
[18]
P. Skudlarski, R. T. Constable, and J. C. Gore, “ROC analysis of statistical methods used in functional MRI: individual subjects,” NeuroImage, vol. 9, no. 3, pp. 311–329, 1999.
[19]
R. R. Nandy and D. Cordes, “Novel ROC-type method for testing the efficiency of multivariate statistical methods in fMRI,” Magnetic Resonance in Medicine, vol. 49, no. 6, pp. 1152–1162, 2003.
[20]
R. R. Nandy and D. Cordes, “Novel nonparametric approach to canonical correlation analysis with applications to low CNR functional MRI data,” Magnetic Resonance in Medicine, vol. 50, no. 2, pp. 354–365, 2003.
[21]
R. R. Nandy and D. Cordes, “New approaches to receiver operating characteristic methods in functional magnetic resonance imaging with real data using repeated trials,” Magnetic Resonance in Medicine, vol. 52, no. 6, pp. 1424–1431, 2004.
[22]
R. Nandy and D. Cordes, “A semi-parametric approach to estimate the family-wise error rate in fMRI using resting-state data,” NeuroImage, vol. 34, no. 4, pp. 1562–1576, 2007.
[23]
A. Rencher, Multivariate Statistical Inference and Applications, John Wiley & Sons, New York, NY, USA, 1998.
[24]
D. Cordes, M. Jin, T. Curran, and R. Nandy, “Optimizing the performance of local canonical correlation analysis in fMRI using spatial constraints,” Human Brain Mapping. In press.
[25]
D. Prichard and J. Theiler, “Generating surrogate data for time series with several simultaneously measured variables,” Physical Review Letters, vol. 73, no. 7, pp. 951–954, 1994.
[26]
E. Bullmore, C. Long, J. Suckling et al., “Colored noise and computational inference in neurophysiological (fMRI) time series analysis: resampling methods in time and wavelet domains,” Human Brain Mapping, vol. 12, no. 2, pp. 61–78, 2001.
[27]
M. Breakspear, M. J. Brammer, E. T. Bullmore, P. Das, and L. M. Williams, “Spatiotemporal wavelet resampling for functional neuroimaging data,” Human Brain Mapping, vol. 23, no. 1, pp. 1–25, 2004.
[28]
E. Bullmore, M. Brammer, S. C. R. Williams et al., “Statistical methods of estimation and inference for functional MR image analysis,” Magnetic Resonance in Medicine, vol. 35, no. 2, pp. 261–277, 1996.
[29]
J. J. Locascio, P. J. Jennings, C. I. Moore, and S. Corkin, “Time series analysis in the time domain and resampling methods for studies of functional magnetic resonance brain imaging,” Human Brain Mapping, vol. 5, no. 3, pp. 168–193, 1997.
[30]
R. Nandy, C. Green, and D. Cordes, “Nonparametric analysis of fMRI data using bootstrap in autoregression,” Proceedings ISMRM, vol. 10, p. 1438, 2002.
[31]
O. Friman and C. F. Westin, “Resampling fMRI time series,” NeuroImage, vol. 25, no. 3, pp. 859–867, 2005.
[32]
A. R. Laird, B. P. Rogers, and M. E. Meyerand, “Comparison of Fourier and wavelet resampling methods,” Magnetic Resonance in Medicine, vol. 51, no. 2, pp. 418–422, 2004.
[33]
M. Jin and D. Cordes, “Validation of resampling methods for fMRI data,” Neuroimage, vol. 41, p. S525, 2008.
[34]
R. Pyke, “Spacings (with discussion),” Journal of the Royal Statistical Society: Series B, vol. 27, pp. 395–449, 1965.
[35]
I. Weissman, “Estimation of parameters and larger quantiles based on the k largest observations,” Journal of the American Statistical Association, vol. 73, pp. 812–815, 1978.
[36]
U. E. Ruttimann, M. Unser, R. R. Rawlings et al., “Statistical analysis of functional MRI data in the wavelet domain,” IEEE Transactions on Medical Imaging, vol. 17, no. 2, pp. 142–154, 1998.
[37]
G. A. Hossein-Zadeh, B. A. Ardekani, and H. Soltanian-Zadeh, “Activation detection in fMRI using a maximum energy ratio statistic obtained by adaptive spatial filtering,” IEEE Transactions on Medical Imaging, vol. 22, no. 7, pp. 795–805, 2003.
[38]
K. Tabelow, J. Polzehl, H. U. Voss, and V. Spokoiny, “Analyzing fMRI experiments with structural adaptive smoothing procedures,” NeuroImage, vol. 33, no. 1, pp. 55–62, 2006.
[39]
W. D. Penny, N. J. Trujillo-Barreto, and K. J. Friston, “Bayesian fMRI time series analysis with spatial priors,” NeuroImage, vol. 24, no. 2, pp. 350–362, 2005.
[40]
G. Flandin and W. D. Penny, “Bayesian fMRI data analysis with sparse spatial basis function priors,” NeuroImage, vol. 34, no. 3, pp. 1108–1125, 2007.
[41]
J. R. Petrella, L. M. Shah, K. M. Harris et al., “Preoperative functional MR imaging localization of language and motor areas: effect on therapeutic decision making in patients with potentially resectable brain tumors,” Radiology, vol. 240, no. 3, pp. 793–802, 2006.
[42]
M. M. Zeineh, S. A. Engel, and S. Y. Bookheimer, “Application of cortical unfolding techniques to functional MRI of the human hippocampal region,” NeuroImage, vol. 11, no. 6 I, pp. 668–683, 2000.
