A family of locally constrained CCA models for detecting activation patterns in fMRI

Xiaowei Zhuang, Zhengshi Yang, Tim Curran, Richard Byrd, Rajesh Ranjan Nandy, Dietmar Cordes

Research output: Contribution to journalArticleResearchpeer-review

6 Citations (Scopus)

Abstract

Canonical correlation analysis (CCA) has been used in Functional Magnetic Resonance Imaging (fMRI) for improved detection of activation by incorporating time series from multiple voxels in a local neighborhood. To improve the specificity of local CCA methods, spatial constraints were previously proposed. In this study, constraints are generalized by introducing a family model of spatial constraints for CCA to further increase both sensitivity and specificity in fMRI activation detection. The proposed locally-constrained CCA (cCCA) model is formulated in terms of a multivariate constrained optimization problem and solved efficiently with numerical optimization techniques. To evaluate the performance of this cCCA model, simulated data are generated with a Signal-To-Noise Ratio of 0.25, which is realistic to the noise level contained in episodic memory fMRI data. Receiver operating characteristic (ROC) methods are used to compare the performance of different models. The cCCA model with optimum parameters (called optimum-cCCA) obtains the largest area under the ROC curve. Furthermore, a novel validation method is proposed to validate the selected optimum-cCCA parameters based on ROC from simulated data and real fMRI data. Results for optimum-cCCA are then compared with conventional fMRI analysis methods using data from an episodic memory task. Wavelet-resampled resting-state data are used to obtain the null distribution of activation. For simulated data, accuracy in detecting activation increases for the optimum-cCCA model by about 43% as compared to the single voxel analysis with comparable Gaussian smoothing. Results from the real fMRI data set indicate a significant increase in activation detection, particularly in hippocampus, para-hippocampal area and nearby medial temporal lobe regions with the proposed method.

Original languageEnglish
Pages (from-to)63-84
Number of pages22
JournalNeuroImage
Volume149
DOIs
StatePublished - 1 Apr 2017

Fingerprint

Magnetic Resonance Imaging
ROC Curve
Episodic Memory
Temporal Lobe
Spatial Analysis
Signal-To-Noise Ratio
Noise
Hippocampus
Sensitivity and Specificity

Keywords

  • Constrained canonical correlation analysis (cCCA)
  • Episodic memory task
  • Functional Magnetic Resonance Imaging (fMRI)
  • Multivariate analysis
  • Numerical optimization

Cite this

Zhuang, Xiaowei ; Yang, Zhengshi ; Curran, Tim ; Byrd, Richard ; Nandy, Rajesh Ranjan ; Cordes, Dietmar. / A family of locally constrained CCA models for detecting activation patterns in fMRI. In: NeuroImage. 2017 ; Vol. 149. pp. 63-84.
@article{c80498e2c7be46e39a0775a8a1cb9b3e,
title = "A family of locally constrained CCA models for detecting activation patterns in fMRI",
abstract = "Canonical correlation analysis (CCA) has been used in Functional Magnetic Resonance Imaging (fMRI) for improved detection of activation by incorporating time series from multiple voxels in a local neighborhood. To improve the specificity of local CCA methods, spatial constraints were previously proposed. In this study, constraints are generalized by introducing a family model of spatial constraints for CCA to further increase both sensitivity and specificity in fMRI activation detection. The proposed locally-constrained CCA (cCCA) model is formulated in terms of a multivariate constrained optimization problem and solved efficiently with numerical optimization techniques. To evaluate the performance of this cCCA model, simulated data are generated with a Signal-To-Noise Ratio of 0.25, which is realistic to the noise level contained in episodic memory fMRI data. Receiver operating characteristic (ROC) methods are used to compare the performance of different models. The cCCA model with optimum parameters (called optimum-cCCA) obtains the largest area under the ROC curve. Furthermore, a novel validation method is proposed to validate the selected optimum-cCCA parameters based on ROC from simulated data and real fMRI data. Results for optimum-cCCA are then compared with conventional fMRI analysis methods using data from an episodic memory task. Wavelet-resampled resting-state data are used to obtain the null distribution of activation. For simulated data, accuracy in detecting activation increases for the optimum-cCCA model by about 43{\%} as compared to the single voxel analysis with comparable Gaussian smoothing. Results from the real fMRI data set indicate a significant increase in activation detection, particularly in hippocampus, para-hippocampal area and nearby medial temporal lobe regions with the proposed method.",
keywords = "Constrained canonical correlation analysis (cCCA), Episodic memory task, Functional Magnetic Resonance Imaging (fMRI), Multivariate analysis, Numerical optimization",
author = "Xiaowei Zhuang and Zhengshi Yang and Tim Curran and Richard Byrd and Nandy, {Rajesh Ranjan} and Dietmar Cordes",
year = "2017",
month = "4",
day = "1",
doi = "10.1016/j.neuroimage.2016.12.081",
language = "English",
volume = "149",
pages = "63--84",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

A family of locally constrained CCA models for detecting activation patterns in fMRI. / Zhuang, Xiaowei; Yang, Zhengshi; Curran, Tim; Byrd, Richard; Nandy, Rajesh Ranjan; Cordes, Dietmar.

In: NeuroImage, Vol. 149, 01.04.2017, p. 63-84.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A family of locally constrained CCA models for detecting activation patterns in fMRI

AU - Zhuang, Xiaowei

AU - Yang, Zhengshi

AU - Curran, Tim

AU - Byrd, Richard

AU - Nandy, Rajesh Ranjan

AU - Cordes, Dietmar

PY - 2017/4/1

Y1 - 2017/4/1

N2 - Canonical correlation analysis (CCA) has been used in Functional Magnetic Resonance Imaging (fMRI) for improved detection of activation by incorporating time series from multiple voxels in a local neighborhood. To improve the specificity of local CCA methods, spatial constraints were previously proposed. In this study, constraints are generalized by introducing a family model of spatial constraints for CCA to further increase both sensitivity and specificity in fMRI activation detection. The proposed locally-constrained CCA (cCCA) model is formulated in terms of a multivariate constrained optimization problem and solved efficiently with numerical optimization techniques. To evaluate the performance of this cCCA model, simulated data are generated with a Signal-To-Noise Ratio of 0.25, which is realistic to the noise level contained in episodic memory fMRI data. Receiver operating characteristic (ROC) methods are used to compare the performance of different models. The cCCA model with optimum parameters (called optimum-cCCA) obtains the largest area under the ROC curve. Furthermore, a novel validation method is proposed to validate the selected optimum-cCCA parameters based on ROC from simulated data and real fMRI data. Results for optimum-cCCA are then compared with conventional fMRI analysis methods using data from an episodic memory task. Wavelet-resampled resting-state data are used to obtain the null distribution of activation. For simulated data, accuracy in detecting activation increases for the optimum-cCCA model by about 43% as compared to the single voxel analysis with comparable Gaussian smoothing. Results from the real fMRI data set indicate a significant increase in activation detection, particularly in hippocampus, para-hippocampal area and nearby medial temporal lobe regions with the proposed method.

AB - Canonical correlation analysis (CCA) has been used in Functional Magnetic Resonance Imaging (fMRI) for improved detection of activation by incorporating time series from multiple voxels in a local neighborhood. To improve the specificity of local CCA methods, spatial constraints were previously proposed. In this study, constraints are generalized by introducing a family model of spatial constraints for CCA to further increase both sensitivity and specificity in fMRI activation detection. The proposed locally-constrained CCA (cCCA) model is formulated in terms of a multivariate constrained optimization problem and solved efficiently with numerical optimization techniques. To evaluate the performance of this cCCA model, simulated data are generated with a Signal-To-Noise Ratio of 0.25, which is realistic to the noise level contained in episodic memory fMRI data. Receiver operating characteristic (ROC) methods are used to compare the performance of different models. The cCCA model with optimum parameters (called optimum-cCCA) obtains the largest area under the ROC curve. Furthermore, a novel validation method is proposed to validate the selected optimum-cCCA parameters based on ROC from simulated data and real fMRI data. Results for optimum-cCCA are then compared with conventional fMRI analysis methods using data from an episodic memory task. Wavelet-resampled resting-state data are used to obtain the null distribution of activation. For simulated data, accuracy in detecting activation increases for the optimum-cCCA model by about 43% as compared to the single voxel analysis with comparable Gaussian smoothing. Results from the real fMRI data set indicate a significant increase in activation detection, particularly in hippocampus, para-hippocampal area and nearby medial temporal lobe regions with the proposed method.

KW - Constrained canonical correlation analysis (cCCA)

KW - Episodic memory task

KW - Functional Magnetic Resonance Imaging (fMRI)

KW - Multivariate analysis

KW - Numerical optimization

UR - http://www.scopus.com/inward/record.url?scp=85011407810&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2016.12.081

DO - 10.1016/j.neuroimage.2016.12.081

M3 - Article

VL - 149

SP - 63

EP - 84

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -