3D spatially-adaptive canonical correlation analysis: Local and global methods

Zhengshi Yang, Xiaowei Zhuang, Karthik Sreenivasan, Virendra Mishra, Tim Curran, Richard Byrd, Rajesh Ranjan Nandy, Dietmar Cordes

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Local spatially-adaptive canonical correlation analysis (local CCA) with spatial constraints has been introduced to fMRI multivariate analysis for improved modeling of activation patterns. However, current algorithms require complicated spatial constraints that have only been applied to 2D local neighborhoods because the computational time would be exponentially increased if the same method is applied to 3D spatial neighborhoods. In this study, an efficient and accurate line search sequential quadratic programming (SQP) algorithm has been developed to efficiently solve the 3D local CCA problem with spatial constraints. In addition, a spatially-adaptive kernel CCA (KCCA) method is proposed to increase accuracy of fMRI activation maps. With oriented 3D spatial filters anisotropic shapes can be estimated during the KCCA analysis of fMRI time courses. These filters are orientation-adaptive leading to rotational invariance to better match arbitrary oriented fMRI activation patterns, resulting in improved sensitivity of activation detection while significantly reducing spatial blurring artifacts. The kernel method in its basic form does not require any spatial constraints and analyzes the whole-brain fMRI time series to construct an activation map. Finally, we have developed a penalized kernel CCA model that involves spatial low-pass filter constraints to increase the specificity of the method. The kernel CCA methods are compared with the standard univariate method and with two different local CCA methods that were solved by the SQP algorithm. Results show that SQP is the most efficient algorithm to solve the local constrained CCA problem, and the proposed kernel CCA methods outperformed univariate and local CCA methods in detecting activations for both simulated and real fMRI episodic memory data.

Original languageEnglish
Pages (from-to)240-255
Number of pages16
JournalNeuroImage
Volume169
DOIs
StatePublished - 1 Apr 2018

Fingerprint

Magnetic Resonance Imaging
Spatial Analysis
Episodic Memory
Artifacts
Multivariate Analysis
Brain

Keywords

  • Constrained canonical correlation analysis
  • Kernel canonical correlation analysis
  • Multivariate analysis
  • Orientation filters
  • Spatial filtering
  • fMRI

Cite this

Yang, Z., Zhuang, X., Sreenivasan, K., Mishra, V., Curran, T., Byrd, R., ... Cordes, D. (2018). 3D spatially-adaptive canonical correlation analysis: Local and global methods. NeuroImage, 169, 240-255. https://doi.org/10.1016/j.neuroimage.2017.12.025
Yang, Zhengshi ; Zhuang, Xiaowei ; Sreenivasan, Karthik ; Mishra, Virendra ; Curran, Tim ; Byrd, Richard ; Nandy, Rajesh Ranjan ; Cordes, Dietmar. / 3D spatially-adaptive canonical correlation analysis : Local and global methods. In: NeuroImage. 2018 ; Vol. 169. pp. 240-255.
@article{5fc20723848f4a5cad2c3e2b24e345bd,
title = "3D spatially-adaptive canonical correlation analysis: Local and global methods",
abstract = "Local spatially-adaptive canonical correlation analysis (local CCA) with spatial constraints has been introduced to fMRI multivariate analysis for improved modeling of activation patterns. However, current algorithms require complicated spatial constraints that have only been applied to 2D local neighborhoods because the computational time would be exponentially increased if the same method is applied to 3D spatial neighborhoods. In this study, an efficient and accurate line search sequential quadratic programming (SQP) algorithm has been developed to efficiently solve the 3D local CCA problem with spatial constraints. In addition, a spatially-adaptive kernel CCA (KCCA) method is proposed to increase accuracy of fMRI activation maps. With oriented 3D spatial filters anisotropic shapes can be estimated during the KCCA analysis of fMRI time courses. These filters are orientation-adaptive leading to rotational invariance to better match arbitrary oriented fMRI activation patterns, resulting in improved sensitivity of activation detection while significantly reducing spatial blurring artifacts. The kernel method in its basic form does not require any spatial constraints and analyzes the whole-brain fMRI time series to construct an activation map. Finally, we have developed a penalized kernel CCA model that involves spatial low-pass filter constraints to increase the specificity of the method. The kernel CCA methods are compared with the standard univariate method and with two different local CCA methods that were solved by the SQP algorithm. Results show that SQP is the most efficient algorithm to solve the local constrained CCA problem, and the proposed kernel CCA methods outperformed univariate and local CCA methods in detecting activations for both simulated and real fMRI episodic memory data.",
keywords = "Constrained canonical correlation analysis, Kernel canonical correlation analysis, Multivariate analysis, Orientation filters, Spatial filtering, fMRI",
author = "Zhengshi Yang and Xiaowei Zhuang and Karthik Sreenivasan and Virendra Mishra and Tim Curran and Richard Byrd and Nandy, {Rajesh Ranjan} and Dietmar Cordes",
year = "2018",
month = "4",
day = "1",
doi = "10.1016/j.neuroimage.2017.12.025",
language = "English",
volume = "169",
pages = "240--255",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

Yang, Z, Zhuang, X, Sreenivasan, K, Mishra, V, Curran, T, Byrd, R, Nandy, RR & Cordes, D 2018, '3D spatially-adaptive canonical correlation analysis: Local and global methods', NeuroImage, vol. 169, pp. 240-255. https://doi.org/10.1016/j.neuroimage.2017.12.025

3D spatially-adaptive canonical correlation analysis : Local and global methods. / Yang, Zhengshi; Zhuang, Xiaowei; Sreenivasan, Karthik; Mishra, Virendra; Curran, Tim; Byrd, Richard; Nandy, Rajesh Ranjan; Cordes, Dietmar.

In: NeuroImage, Vol. 169, 01.04.2018, p. 240-255.

Research output: Contribution to journalArticle

TY - JOUR

T1 - 3D spatially-adaptive canonical correlation analysis

T2 - Local and global methods

AU - Yang, Zhengshi

AU - Zhuang, Xiaowei

AU - Sreenivasan, Karthik

AU - Mishra, Virendra

AU - Curran, Tim

AU - Byrd, Richard

AU - Nandy, Rajesh Ranjan

AU - Cordes, Dietmar

PY - 2018/4/1

Y1 - 2018/4/1

N2 - Local spatially-adaptive canonical correlation analysis (local CCA) with spatial constraints has been introduced to fMRI multivariate analysis for improved modeling of activation patterns. However, current algorithms require complicated spatial constraints that have only been applied to 2D local neighborhoods because the computational time would be exponentially increased if the same method is applied to 3D spatial neighborhoods. In this study, an efficient and accurate line search sequential quadratic programming (SQP) algorithm has been developed to efficiently solve the 3D local CCA problem with spatial constraints. In addition, a spatially-adaptive kernel CCA (KCCA) method is proposed to increase accuracy of fMRI activation maps. With oriented 3D spatial filters anisotropic shapes can be estimated during the KCCA analysis of fMRI time courses. These filters are orientation-adaptive leading to rotational invariance to better match arbitrary oriented fMRI activation patterns, resulting in improved sensitivity of activation detection while significantly reducing spatial blurring artifacts. The kernel method in its basic form does not require any spatial constraints and analyzes the whole-brain fMRI time series to construct an activation map. Finally, we have developed a penalized kernel CCA model that involves spatial low-pass filter constraints to increase the specificity of the method. The kernel CCA methods are compared with the standard univariate method and with two different local CCA methods that were solved by the SQP algorithm. Results show that SQP is the most efficient algorithm to solve the local constrained CCA problem, and the proposed kernel CCA methods outperformed univariate and local CCA methods in detecting activations for both simulated and real fMRI episodic memory data.

AB - Local spatially-adaptive canonical correlation analysis (local CCA) with spatial constraints has been introduced to fMRI multivariate analysis for improved modeling of activation patterns. However, current algorithms require complicated spatial constraints that have only been applied to 2D local neighborhoods because the computational time would be exponentially increased if the same method is applied to 3D spatial neighborhoods. In this study, an efficient and accurate line search sequential quadratic programming (SQP) algorithm has been developed to efficiently solve the 3D local CCA problem with spatial constraints. In addition, a spatially-adaptive kernel CCA (KCCA) method is proposed to increase accuracy of fMRI activation maps. With oriented 3D spatial filters anisotropic shapes can be estimated during the KCCA analysis of fMRI time courses. These filters are orientation-adaptive leading to rotational invariance to better match arbitrary oriented fMRI activation patterns, resulting in improved sensitivity of activation detection while significantly reducing spatial blurring artifacts. The kernel method in its basic form does not require any spatial constraints and analyzes the whole-brain fMRI time series to construct an activation map. Finally, we have developed a penalized kernel CCA model that involves spatial low-pass filter constraints to increase the specificity of the method. The kernel CCA methods are compared with the standard univariate method and with two different local CCA methods that were solved by the SQP algorithm. Results show that SQP is the most efficient algorithm to solve the local constrained CCA problem, and the proposed kernel CCA methods outperformed univariate and local CCA methods in detecting activations for both simulated and real fMRI episodic memory data.

KW - Constrained canonical correlation analysis

KW - Kernel canonical correlation analysis

KW - Multivariate analysis

KW - Orientation filters

KW - Spatial filtering

KW - fMRI

UR - http://www.scopus.com/inward/record.url?scp=85038845795&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2017.12.025

DO - 10.1016/j.neuroimage.2017.12.025

M3 - Article

C2 - 29248697

AN - SCOPUS:85038845795

VL - 169

SP - 240

EP - 255

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -

Yang Z, Zhuang X, Sreenivasan K, Mishra V, Curran T, Byrd R et al. 3D spatially-adaptive canonical correlation analysis: Local and global methods. NeuroImage. 2018 Apr 1;169:240-255. https://doi.org/10.1016/j.neuroimage.2017.12.025