Shift-Invariant Canonical Polyadic Decomposition of Complex-Valued Multi-Subject fMRI Data With a Phase Sparsity Constraint

Kuang, Li-Dan; Lin, Qiu‐Hua; Gong, Xiaofeng; Cong, Fengyu; Wang, Yu Ping; Calhoun, Vince D.

doi:10.1109/tmi.2019.2936046

Cited by 24 publications

(15 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned above, existing TKD algorithms for fMRI analyses mostly utilize orthogonality constraints to extract FC patterns across time and subjects. These algorithms, however, are unsuitable for extracting common spatial and temporal patterns across subjects due to distinct characteristics such as high-level noise [26]. Motivated by the success of TKD-based image denoising algorithms such as RKCA, we propose to enforce fMRI-specific constraints on Tucker-2 model to analyze three-way multi-subject fMRI data (voxel × time × subject).…”

Section: Introductionmentioning

confidence: 99%

Low-Rank Tucker-2 Model for Multi-Subject fMRI Data Decomposition With Spatial Sparsity Constraint

Yue

Lin

Kuang

et al. 2022

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

Tucker decomposition can provide an intuitive summary to understand brain function by decomposing multisubject fMRI data into a core tensor and multiple factor matrices, and was mostly used to extract functional connectivity patterns across time/subjects using orthogonality constraints. However, these algorithms are unsuitable for extracting common spatial and temporal patterns across subjects due to distinct characteristics such as high-level noise. Motivated by a successful application of Tucker decomposition to image denoising and the intrinsic sparsity of spatial activations in fMRI, we propose a low-rank Tucker-2 model with spatial sparsity constraint to analyze multisubject fMRI data. More precisely, we propose to impose a sparsity constraint on spatial maps by using an p  norm (0

show abstract

Section: Introductionmentioning

confidence: 99%

Low-Rank Tucker-2 Model for Multi-Subject fMRI Data Decomposition With Spatial Sparsity Constraint

Yue

Lin

Kuang

et al. 2022

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…Qiu et al utilized SSP to identify more significant variance changes and higher sensitivity to the spatial differences between HCs and SZs, compared to the magnitude of spatial activations [6]. Kuang et al proposed to impose an SSP sparsity constraint on the complex-valued shared spatial maps to improve separation performance of tensor decomposition [7].…”

Section: Introductionmentioning

confidence: 99%

Sparse Representation of Complex-Valued fMRI Data Based on Hard Thresholding of Spatial Source Phase

Song

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Spatial source phase (SSP), derived from complex-valued functional magnetic resonance imaging (fMRI) data by datadriven methods, has unique capacity of identifying blood oxygenation-level dependent (BOLD)-related voxels from noisy voxels regardless of their amplitudes. However, the use of SSP constraint in sparse representation algorithms have rarely been studied. This study proposes a sparse representation method using SSP hard thresholding to achieve the sparsity of spatial components, enabling the use of initially complex-valued fMRI data and retaining the brain information embedded in noisy voxels and weak BOLDrelated voxels with small phase values. Rank-1 matrix estimation is applied to sequentially update dictionary atoms and corresponding spatial components, followed by hard thresholding on spatial components based on SSP. The proposed method is evaluated using both simulated and experimental complex-valued data. The results show that the proposed method yields better performance than a complexvalued dictionary learning algorithm when using initially acquired complex-valued task-related fMRI data.

show abstract

“…Multi-subject fMRI data can provide shared and/or individual spatial and temporal components for understanding brain functions and mental disorders via blind source separation (BSS) [1]- [6]. Among others, group independent component analysis (GICA) performs ICA on concatenated singlesubject fMRI data along space or time to obtain either shared time courses (TCs) and individual spatial maps (SMs) or shared SMs and individual TCs [1] [2].…”

Section: Introductionmentioning

confidence: 99%

“…Independent vector analysis (IVA) carries out joint ICA of single-subject fMRI data to obtain individual SMs and TCs from a group-level viewpoint [3][4]. Compared with GICA and IVA, tensor decomposition makes full use of structural information of multi-subject fMRI data to provide shared SMs and TCs [5] [6]. Canonical polyadic decomposition (CPD) and Tucker decomposition (TKD) and are two commonly utilized tensor decomposition algorithms.…”

Section: Introductionmentioning

confidence: 99%

Tucker Decomposition for Extracting Shared and Individual Spatial Maps from Multi-Subject Resting-State fMRI Data

Yue

Lin

Kuang

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Tucker decomposition (TKD) has been utilized to identify functional connectivity patterns using processed fMRI data, but seldom focuses on originally acquired fMRI data. This study proposes to decompose multi-subject fMRI data in a natural three-way of voxel × time × subject via TKD. Different from existing tensor decomposition algorithms such as canonical polyadic decomposition (CPD) for extracting shared spatial maps (SMs), we propose to extract both shared and individual SMs by exploring spatial-temporal-subject relationship contained in the core tensor. We test the proposed method using multi-subject resting-state fMRI data with comparison to CPD for evaluating shared SMs and independent vector analysis (IVA) for assessing individual SMs under different model orders. The results show that the proposed method yields better and more robust shared SMs than CPD and more consistent individual SMs than IVA, indicating the potential of TKD in providing group and individual brain networks in a high-dimensional coupling way.

show abstract

Shift-Invariant Canonical Polyadic Decomposition of Complex-Valued Multi-Subject fMRI Data With a Phase Sparsity Constraint

Abstract: This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details.

Cited by 24 publications

References 40 publications

Low-Rank Tucker-2 Model for Multi-Subject fMRI Data Decomposition With Spatial Sparsity Constraint

Low-Rank Tucker-2 Model for Multi-Subject fMRI Data Decomposition With Spatial Sparsity Constraint

Sparse Representation of Complex-Valued fMRI Data Based on Hard Thresholding of Spatial Source Phase

Tucker Decomposition for Extracting Shared and Individual Spatial Maps from Multi-Subject Resting-State fMRI Data

Contact Info

Product

Resources

About