Tensor Regression with Applications in Neuroimaging Data Analysis

Abstract

Classical regression methods treat covariates as a vector and estimate a corresponding vector of regression coefficients. Modern applications in medical imaging generate covariates of more complex form such as multidimensional arrays (tensors). Traditional statistical and computational methods are proving insufficient for analysis of these high-throughput data due to their ultrahigh dimensionality as well as complex structure. In this article, we propose a new family of tensor regression models that efficiently exploit the special structure of tensor covariates. Under this framework, ultrahigh dimensionality is reduced to a manageable level, resulting in efficient estimation and prediction. A fast and highly scalable estimation algorithm is proposed for maximum likelihood estimation and its associated asymptotic properties are studied. Effectiveness of the new methods is demonstrated on both synthetic and real MRI imaging data.

27 pages, 4 figures

References in corpus (3)

• The Statistical Analysis of fMRI Data
• On dimension folding of matrix- or array-valued statistical objects
• Path Following and Empirical Bayes Model Selection for Sparse Regression

Cited by in corpus (120)

• Tensor Networks for Dimensionality Reduction and Large-Scale Optimizations. Part 2 Applications and Future Perspectives
• Regularized Matrix Regression
• Tensor Methods in Computer Vision and Deep Learning
• Tensor-Based Classifiers for Hyperspectral Data Analysis
• Tensor-on-tensor regression
• Big Data and Reliability Applications: The Complexity Dimension
• Tensor Regression Networks
• STORE: Sparse Tensor Response Regression and Neuroimaging Analysis
• A novel approach for fusion of heterogeneous sources of data
• Structured Point Cloud Data Analysis via Regularized Tensor Regression for Process Modeling and Optimization
• Network classification with applications to brain connectomics
• Independent component analysis for tensor-valued data
• Improving Efficiency in Convolutional Neural Network with Multilinear Filters
• Provable Convex Co-clustering of Tensors
• Wavelet-domain regression and predictive inference in psychiatric neuroimaging
• Symmetric Bilinear Regression for Signal Subgraph Estimation
• Rank Determination in Tensor Factor Model
• Tensor Regression Using Low-rank and Sparse Tucker Decompositions
• Learning from Multiway Data: Simple and Efficient Tensor Regression
• Deep Kronecker Network
• Tensor-based Nonlinear Classifier for High-Order Data Analysis
• Tucker Tensor Regression and Neuroimaging Analysis
• Supervised multiway factorization
• Doubly Decomposing Nonparametric Tensor Regression
• Reduced-Rank Tensor-on-Tensor Regression and Tensor-variate Analysis of Variance
• Canonical Polyadic Decomposition with Auxiliary Information for Brain Computer Interface
• Scaling and Scalability: Provable Nonconvex Low-Rank Tensor Estimation from Incomplete Measurements
• Discriminating sample groups with multi-way data
• Broadcasted Nonparametric Tensor Regression
• Tensor Factor Model Estimation by Iterative Projection
• Robust Tensor CUR Decompositions: Rapid Low-Tucker-Rank Tensor Recovery with Sparse Corruption
• Learning from Binary Multiway Data: Probabilistic Tensor Decomposition and its Statistical Optimality
• The Impact of Model Assumptions in Scalar-on-Image Regression
• Efficient Structure-preserving Support Tensor Train Machine
• On the Convergence of Asynchronous Parallel Iteration with Unbounded Delays
• High-Dimensional Low-Rank Tensor Autoregressive Time Series Modeling
• Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis
• Exact Clustering in Tensor Block Model: Statistical Optimality and Computational Limit
• Supervised tensor decomposition with features on multiple modes
• Tensor Generalized Estimating Equations for Longitudinal Imaging Analysis
• A Sharp Blockwise Tensor Perturbation Bound for Orthogonal Iteration
• High-dimensional vector autoregressive time series modeling via tensor decomposition
• Low-rank Tensor Estimation via Riemannian Gauss-Newton: Statistical Optimality and Second-Order Convergence
• CP Degeneracy in Tensor Regression
• Partially Observed Dynamic Tensor Response Regression
• DuSK: A Dual Structure-preserving Kernel for Supervised Tensor Learning with Applications to Neuroimages
• Power Iteration for Tensor PCA
• A Statistically Identifiable Model for Tensor-Valued Gaussian Random Variables
• Decomposing Network Influence: Social Influence Regression
• Autism Spectrum Disorder Classification using Graph Kernels on Multidimensional Time Series
• An Alternative Approach to Functional Linear Partial Quantile Regression
• Adaptive Learning of Tensor Network Structures
• Proximal MCMC for Bayesian Inference of Constrained and Regularized Estimation
• A low-rank based estimation-testing procedure for matrix-covariate regression
• Higher order Matching Pursuit for Low Rank Tensor Learning
• Jointly Modeling and Clustering Tensors in High Dimensions
• Cross: Efficient Low-rank Tensor Completion
• Adaptive Higher-order Spectral Estimators
• Tensorized Random Projections
• Tensor denoising and completion based on ordinal observations
• Provable Sparse Tensor Decomposition
• Bayesian Nonlinear Tensor Regression with Functional Fused Elastic Net Prior
• A review of heterogeneous data mining for brain disorders
• Bayesian Tensor Regression
• Autoregressive Models for Matrix-Valued Time Series
• Tensor SVD: Statistical and Computational Limits
• Scalable Spatiotemporally Varying Coefficient Modelling with Bayesian Kernelized Tensor Regression
• Efficient Multidimensional Functional Data Analysis Using Marginal Product Basis Systems
• Parsimonious Tensor Response Regression
• TULIP: A Toolbox for Linear Discriminant Analysis with Penalties
• Long Random Matrices and Tensor Unfolding
• Fast Low-Rank Tensor Decomposition by Ridge Leverage Score Sampling
• AutoGFI: Streamlined Generalized Fiducial Inference for Modern Inference Problems in Models with Additive Errors
• Bayesian Time-Varying Tensor Vector Autoregressive Models for Dynamic Effective Connectivity
• LOCUS: A Novel Decomposition Method for Brain Network Connectivity Matrices using Low-rank Structure with Uniform Sparsity
• Sparse Tensor Additive Regression
• DBSCAN of Multi-Slice Clustering for Third-Order Tensors
• SUM: Suboptimal Unitary Multi-task Learning Framework for Spatiotemporal Data Prediction
• Dynamic Tensor Clustering
• Graph-Fused Multivariate Regression via Total Variation Regularization
• Stable and predictive functional domain selection with application to brain images
• Learning Signal Subgraphs from Longitudinal Brain Networks with Symmetric Bilinear Logistic Regression
• Dynamic Matrix Recovery
• Sparse Symmetric Tensor Regression for Functional Connectivity Analysis
• A Doubly-Enhanced EM Algorithm for Model-Based Tensor Clustering
• Maximum Regularized Likelihood Estimators: A General Prediction Theory and Applications
• Personalized Tucker Decomposition: Modeling Commonality and Peculiarity on Tensor Data
• PASTA: A Parallel Sparse Tensor Algorithm Benchmark Suite
• Low-Rank Tensor Regression for X-Ray Tomography
• Image-Based Prognostics Using Penalized Tensor Regression
• Statistical Challenges in Modeling Big Brain Signals
• ISLET: Fast and Optimal Low-rank Tensor Regression via Importance Sketching
• Regularized Kaczmarz Algorithms for Tensor Recovery
• TEC: Tensor Ensemble Classifier for Big Data
• Tensor Convolutional Sparse Coding with Low-Rank activations, an application to EEG analysis
• An Augmented Regression Model for Tensors with Missing Values
• Fast and Scalable Estimator for Sparse and Unit-Rank Higher-Order Regression Models
• Matrix-Variate Regressions and Envelope Models
• A Sparsity Inducing Nuclear-Norm Estimator (SpINNEr) for Matrix-Variate Regression in Brain Connectivity Analysis
• Segmenting High-dimensional Matrix-valued Time Series via Sequential Transformations
• Supervised Multiscale Dimension Reduction for Spatial Interaction Networks
• Theoretical and Experimental Analyses of Tensor-Based Regression and Classification
• Common Mode Patterns for Supervised Tensor Subspace Learning
• Individualized Multilayer Tensor Learning with An Application in Imaging Analysis
• Soft Tensor Regression
• Multivariate Convolutional Sparse Coding with Low Rank Tensor
• Local-Aggregate Modeling for Big-Data via Distributed Optimization: Applications to Neuroimaging
• Sturm: Sparse Tubal-Regularized Multilinear Regression for fMRI
• Auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
• Sparse and Low-Rank High-Order Tensor Regression via Parallel Proximal Method
• Multivariate Temporal Point Process Regression
• Covariate-Adjusted Tensor Classification in High-Dimensions
• Geometric All-Way Boolean Tensor Decomposition
• Tensor Mixed Effects Model with Applications in Nanomanufacturing Inspection
• High-Dimensional Uncertainty Quantification via Tensor Regression with Rank Determination and Adaptive Sampling
• Multi-task Learning with High-Dimensional Noisy Images
• Rademacher Random Projections with Tensor Networks
• Latent Functional PARAFAC for modeling multidimensional longitudinal data
• Sparse Logistic Tensor Decomposition for Binary Data
• Multi-Slice Clustering for 3-order Tensor Data