A Projection CCA Method for Effective fMRI Data Analysis

Abstract

Objective: Canonical correlation analysis (CCA) is a data driven method that has been successfully used in functional magnetic resonance imaging (fMRI) data analysis. Standard CCA extracts meaningful information from a pair of data sets by seeking pairs of linear combinations from two sets of variables with maximum pairwise correlation. So far, however, this method has been used without incorporating prior information available for fMRI data. In this paper, we address this issue by proposing a new CCA method named pCCA (for projection CCA). Methods: The proposed method is obtained by projection onto a set of basis vectors that better characterize temporal information in the fMRI data set. A methodology is presented to describe the basis selection process using discrete cosine transform (DCT) basis functions. Employing DCT guides the estimated canonical variates, yielding a more computationally efficient CCA procedure. Results: The performance gain of the proposed pCCA algorithm over standard and regularized CCA is illustrated on both simulated and real fMRI datasets from resting state, block paradigm task-related and event-related experiments. The results show that pCCA can successfully extract the task as well restingstate latent components with increased specificity. Conclusion: In addition to offering a new CCA approach, when applied on fMRI data, the proposed algorithm adapts to variations of brain activity patterns and reveals the functionally connected brain regions. Significance: The proposed method can be seen as a regularized CCA method where regularization is introduced via basis expansion, which has the advantage of enforcing smoothness on canonical components.

UPLC-Q-Exactive Orbitrap-MS-Based Untargeted Lipidomic Analysis of Lipid Molecular Species in Spinal Cords from Different Domesticated Animals

“…CCA is a well-established multivariate statistical algorithm used to explore the correlation between multivariate datasets and to identify the most relevant features [19,20]. This analytical method has been widely applied in pattern recognition and machine learning.…”

Efficient Blind Source Separation Method for fMRI Using Autoencoder and Spatiotemporal Sparsity Constraints

“…Instead, the data-driven approaches offer a promising alternative due to their adaptability to both task-based activation detection and resting-state functional connectivity analysis [4], [5]. In this regard, some mmBSS methods owing to their computational efficiency, have been very consequential for fMRI studies [6], [7], [8], [9], [10], [11], [12].…”

Sparse Group Bases for Multisubject fMRI Data

“…An obvious disadvantage of this approach is its inability to account for any unpredictable experimental variance, such as HRF variability across subjects [6]. In contrast, the flexibility of data-driven methods such as independent component analysis (ICA) [7], principal component analysis (PCA) [8], canonical correlation analysis (CCA) [9], and their variants [10]- [15] to adapt to individual hemodynamics across subjects and different functional networks by learning underlying trends from the data make them applicable to both TB activation detection and RS functional connectivity analysis. Therefore, these methods have been extensively adapted to fMRI data over the last three decades.…”