A Unified Estimation Framework for State-Related Changes in Effective Brain Connectivity

Samdin, S. Balqis; Ting, Chee-Ming; Ombao, Hernando; Salleh, Shaharuddin

doi:10.1109/tbme.2016.2580738

Cited by 48 publications

(38 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This renders the factor process piecewise stationary, a special form of departure from stationarity. Moreover, compared to using a SVAR model directly on y t as in [15], the specification of (4) allows us to detect the change-points of the high-dimensional dependence structure based on a small number of factor series. We denote the model parameters θ = {θ j = (Φ…”

Section: A the Factor Modelmentioning

confidence: 99%

“…It can be determined by a data-driven approach via estimating the number of clusters in time-variant VAR (TV-VAR) coefficients, using some cluster validity indexes e.g. the variance ratio criterion, as in [15].…”

Section: ) Choice Of Number Of States Kmentioning

confidence: 99%

“…The factor loading matrix Q in H and the noise covariance Σ (common to all regimes) are fixed with the PCA estimates from Step 1, and not updated by EM algorithm. Note that here the regime estimation is done based on the state equation of lowdimensional factors, instead of the high-dimensional latent signals in [15]. This will lead to substantial computational reduction and better estimates of the subspace parameters.…”

Section: ) Choice Of Number Of States Kmentioning

confidence: 99%

“…Our proposed SVAR approach has more advantages than the K-means clustering of time-variant VAR coefficients, as discussed in [15]. First, our approach provides a unified parametric framework for estimating both the dynamic states and connectivity graphs simultaneously, rather than the separate steps of sliding-window connectivity analysis followed by regime clustering.…”

Section: [J]mentioning

confidence: 99%

“…It enables a reliable and computationallyefficient estimation of the regime change-points and the massive dependence measures associated with each regime. The proposed method integrates our two previous approaches with substantial enhancements: (1) It extends the SVAR model [15] by specifying the regime-switching in the lower-dimensional factor subspace rather than the observation space directly, which improves scalability for high-dimensional inference. (2) It generalizes the factor VAR model with time-invariant parameters [16] by introducing nonstationarity in the factor structure, which enables estimation of dynamic connectivity.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models

Ting

Ombao

Samdin

et al. 2018

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

Recent studies on analyzing dynamic brain connectivity rely on sliding-window analysis or timevarying coefficient models which are unable to capture both smooth and abrupt changes simultaneously. Emerging evidence suggests state-related changes in brain connectivity where dependence structure alternates between a finite number of latent states or regimes. Another challenge is inference of full-brain networks with large number of nodes. We employ a Markov-switching dynamic factor model in which the state-driven time-varying connectivity regimes of high-dimensional fMRI data are characterized by lower-dimensional common latent factors, following a regime-switching process. It enables a reliable, data-adaptive estimation of change-points of connectivity regimes and the massive dependencies associated with each regime. We consider the switching VAR to quantity the dynamic effective connectivity. We propose a three-step estimation procedure: (1) extracting the factors using principal component analysis (PCA) and (2) identifying dynamic connectivity states using the factor-based switching vector autoregressive (VAR) models in a state-space formulation using Kalman filter and expectationmaximization (EM) algorithm, and (3) constructing the high-dimensional connectivity metrics for each state based on subspace estimates. Simulation results show that our proposed estimator outperforms the K-means clustering of time-windowed coefficients, providing more accurate estimation of regime dynamics and connectivity metrics in high-dimensional settings. Applications to analyzing restingstate fMRI data identify dynamic changes in brain states during rest, and reveal distinct directed connectivity patterns and modular organization in resting-state networks across different states.

show abstract

Section: A the Factor Modelmentioning

confidence: 99%

Section: ) Choice Of Number Of States Kmentioning

confidence: 99%

Section: ) Choice Of Number Of States Kmentioning

confidence: 99%

Section: [J]mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models

Ting

Ombao

Samdin

et al. 2018

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

Bayesian Methods in Brain Networks

Durante

Guindani

2020

Wiley StatsRef: Statistics Reference Online

View full text Add to dashboard Cite

The recent developments in resonance imaging technologies have allowed growing access to a wide variety of complex information on brain functioning. Regardless of the several types of data modalities which are now available, a fundamental interest in the neurosciences is in conducting inference on brain networks. In this article, we discuss statistical methods for brain networks with a specific focus on the Bayesian approach. Due to its ability to allow careful uncertainty quantification, borrowing of information, and inclusion of experts knowledge, the Bayesian paradigm can provide a particularly effective tool in the neuroscience studies.

show abstract

Bayesian vector autoregressive model for multi‐subject effective connectivity inference using multi‐modal neuroimaging data

Chiang

Guindani

Yeh

et al. 2016

Human Brain Mapping

View full text Add to dashboard Cite

In this article a multi-subject vector autoregressive (VAR) modeling approach was proposed for inference on effective connectivity based on resting-state functional MRI data. Their framework uses a Bayesian variable selection approach to allow for simultaneous inference on effective connectivity at both the subject- and group-level. Furthermore, it accounts for multi-modal data by integrating structural imaging information into the prior model, encouraging effective connectivity between structurally connected regions. They demonstrated through simulation studies that their approach resulted in improved inference on effective connectivity at both the subject- and group-level, compared with currently used methods. It was concluded by illustrating the method on temporal lobe epilepsy data, where resting-state functional MRI and structural MRI were used. Hum Brain Mapp 38:1311-1332, 2017. © 2016 Wiley Periodicals, Inc.

show abstract

A Unified Estimation Framework for State-Related Changes in Effective Brain Connectivity

Abstract: The proposed approach is useful in neuroscience studies that investigate the dynamics of underlying brain states.

Cited by 48 publications

References 45 publications

Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models

Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models

Bayesian Methods in Brain Networks

Bayesian vector autoregressive model for multi‐subject effective connectivity inference using multi‐modal neuroimaging data

Contact Info

Product

Resources

About