Tensorial extensions of independent component analysis for multisubject FMRI analysis

By Christian F. Beckmann, Stephen M. Smith. NeuroImage, 2005

829 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Radiology, Nuclear Medicine and Imaging, Signal Processing

View Paper: https://www.semanticscholar.org/paper/86756e579c78889ca305a50897e2f9070363fe95

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Radiology, Nuclear Medicine and Imaging

In Signal Processing

Citations in the corpus

The Convex Geometry of Linear Inverse Problems (2010, background): “Such problems arise in numerous applications in computer vision and image processing [1], and in neuroscience [4].”
The Statistical Analysis of fMRI Data. (2008, background): “The GIFT approach (Calhoun et al., 2001a) consists of temporally concatenating the data across subjects, and performing ICA decomposition on the resulting data set.”
Advances in functional and structural MR image analysis and implementation as FSL. (2004, methodology): “The PICA methodology is further extended in [5] to the analysis of multi-session/multi-subject FMRI data via a novel exploratory probabilistic tensor-ICA model which provides a tri-linear decomposition and estimates the signal characteristics in the temporal, spatial and subject/session domain.”
Multi-subject Independent Component Analysis of fMRI: A Decade of Intrinsic Networks, Default Mode, and Neurodiagnostic Discovery (2012, neutral): “Three-dimensional tensor decomposition for group and multigroup fMRI data is still being explored, estimating a single spatial, temporal, and subject-specific mode (amplitude parameter) for each component to attempt to capture the multidimensional structure of the data in the estimation stage [51].”
Advances and Pitfalls in the Analysis and Interpretation of Resting-State FMRI Data (2010, neutral): “However, this approach may also be sub-optimal, as it biases towards false-positive fi ndings of group or between-session differences (Calhoun et al., 2001; Beckmann and Smith, 2005).”
Linked independent component analysis for multimodal data fusion (2011, methodology): “This has been used for multi-subject FMRI analysis, with dimensions of voxels×time×subjects (Beckmann and Smith, 2005).”
Comparison of multi-subject ICA methods for analysis of fMRI data (2010, methodology): “One approach is based upon a three-dimensional tensor that has been developed to estimate a single spatial, temporal, and subject-specific ‘‘mode’’ for each component to attempt to capture the multidimensional structure of the data in the estimation stage [Beckmann and Smith, 2005].”
Bayesian analysis of neuroimaging data in FSL (2009, methodology): “…et al., 2006), FLIRT (FMRIB's linear image registration tool; Jenkinson and Smith, 2001), PRELUDE/FUGUE (phase unwarping and MRI unwarping; Jenkinson, 2003), and the " unmixing " algorithm at the heart of the MELODIC ICA-based approach for model-free FMRI analysis (Beckmann and Smith, 2004, 2005).”
Multi-level bootstrap analysis of stable clusters in resting-state fMRI (2009, methodology): “A similar approach (Damoiseaux et al., 2006; Smith et al., 2009) was employed for group-level RSNs generated through tensorial ICA (Beckmann and Smith, 2005).”
Classification and Prediction of Brain Disorders Using Functional Connectivity: Promising but Challenging (2018, methodology): “The tensor probabilistic ICA method stacks the original multi-subject fMRI data along a separate third dimension with a hypothesis that different subjects have common group spatial ICs and time courses but subject specific loading parameters (Beckmann and Smith, 2005; Lee et al., 2008).”
Independent component analysis: recent advances (2013, methodology): “A combination of ICA with PARAFAC for estimation of (5.4) was proposed by Beckmann and Smith (2005), by basically assuming that the S in the PARAFAC model in (5.4) is non-Gaussian like in ICA.”
Fractionating the Default Mode Network: Distinct Contributions of the Ventral and Dorsal Posterior Cingulate Cortex to Cognitive Control (2011, methodology): “Tensor-ICA (Beckmann and Smith, 2005) as implemented in the MELODIC software was first used to define 25 reference network maps common across all four conditions (zero- or two-back, visual or auditory modality).”
Factorization strategies for third-order tensors (2011, background): “Applications involving operations with tensors include chemometrics [41], psychometrics [24], signal processing [9] [27] [39], computer vision [44] [46] [45], data mining [1] [38], graph analysis [21], neuroscience [3] [30] [31], and many more.”
Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion (2011, neutral): “ICA decomposition or regression of motion estimates, to remove spurious motion-related signal from the data (Beckmann and Smith, 2005; Erhardt et al., 2010; Fox et al., 2009; Robinson et al., 2009a; Weissenbacher et al., 2009).”