Gaussian Graphical Models Reveal Inter-Modal and Inter-Regional Conditional Dependencies of Brain Alterations in Alzheimer's Disease

Dyrba, Martin; Mohammadi, Reza; Grothe, Michel J.; Kirste, Thomas; Teipel, Stefan J.

doi:10.3389/fnagi.2020.00099

Cited by 35 publications

(8 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When Y v , v ∈ V are distributed as multivariate Gaussian with positive definite covariance matrix , then G is determined by the zeroes of the inverse covariance matrix and Figure 1. This has been used for inferring undirected PGMs in several applications (Epskamp et al, 2018;Dyrba et al, 2020). In such a case, estimation of G is tantamount to estimation of −1 .…”

Section: Undirected Probabilistic Graphical Modelmentioning

confidence: 99%

See 1 more Smart Citation

Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study

Biswas

Shlizerman

2022

Front. Syst. Neurosci.

View full text Add to dashboard Cite

Representation of brain network interactions is fundamental to the translation of neural structure to brain function. As such, methodologies for mapping neural interactions into structural models, i.e., inference of functional connectome from neural recordings, are key for the study of brain networks. While multiple approaches have been proposed for functional connectomics based on statistical associations between neural activity, association does not necessarily incorporate causation. Additional approaches have been proposed to incorporate aspects of causality to turn functional connectomes into causal functional connectomes, however, these methodologies typically focus on specific aspects of causality. This warrants a systematic statistical framework for causal functional connectomics that defines the foundations of common aspects of causality. Such a framework can assist in contrasting existing approaches and to guide development of further causal methodologies. In this work, we develop such a statistical guide. In particular, we consolidate the notions of associations and representations of neural interaction, i.e., types of neural connectomics, and then describe causal modeling in the statistics literature. We particularly focus on the introduction of directed Markov graphical models as a framework through which we define the Directed Markov Property—an essential criterion for examining the causality of proposed functional connectomes. We demonstrate how based on these notions, a comparative study of several existing approaches for finding causal functional connectivity from neural activity can be conducted. We proceed by providing an outlook ahead regarding the additional properties that future approaches could include to thoroughly address causality.

show abstract

Section: Undirected Probabilistic Graphical Modelmentioning

confidence: 99%

“…This is illustrated in Example 2.1 and Figure 1 . This has been used for inferring undirected PGMs in several applications (Epskamp et al, 2018 ; Dyrba et al, 2020 ). In such a case, estimation of G is tantamount to estimation of Σ −1 .…”

Section: Neural Connectomics: Anatomical and Functionalmentioning

confidence: 99%

Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study

Biswas

Shlizerman

2022

Front. Syst. Neurosci.

View full text Add to dashboard Cite

show abstract

“…This is illustrated in Example 2.1 and Figure 1. This has been used for inferring undirected PGMs in several applications [40,41]. In such a case, estimation of G is tantamount to estimation of Σ −1 .…”

Section: Undirected Probabilistic Graphical Modelmentioning

confidence: 99%

Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study

Biswas,

Shlizerman

2021

Preprint

View full text Add to dashboard Cite

Representation of brain network interactions is fundamental to the translation of neural structure to brain function. As such, methodologies for mapping neural interactions into structural models, i.e., inference of functional connectome from neural recordings, are key for the study of brain networks. While multiple approaches have been proposed for functional connectomics based on statistical associations between neural activity, association does not necessarily incorporate causation. Additional approaches have been proposed to incorporate aspects of causality to turn functional connectomes into causal functional connectomes, however, these methodologies typically focus on specific aspects of causality. This warrants a systematic statistical framework for causal functional connectomics that defines the foundations of common aspects of causality. Such a framework can assist in contrasting existing approaches and to guide development of further causal methodologies. In this work, we develop such a statistical guide. In particular, we consolidate the notions of associations and representations of neural interaction, i.e., types of neural connectomics, and then describe causal modeling in the statistics literature. We particularly focus on the introduction of directed Markov graphical models as a framework through which we define the Directed Markov Property -an essential criterion for examining the causality of proposed functional connectomes. We demonstrate how based on these notions, a comparative study of several existing approaches for finding causal functional connectivity from neural activity can be conducted. We proceed by providing an outlook ahead regarding the additional properties that future approaches could include to thoroughly address causality.Preprint. Under review.

show abstract

“…toxic tau and amyloid-β proteins) spread through communicating cells in the brain (Carbonell et al, 2018; Yasser Iturria-Medina et al, 2014; Vogel et al, 2020)? ; iv ) which multifactorial synergistic (causal) interactions occur at diseased brain regions (Dyrba et al, 2020; Iturria-Medina et al, 2017)? ; and, importantly, v ) how each patient would respond to different therapeutic interventions (Iturria-Medina et al, 2018)?…”

Section: Software Overviewmentioning

confidence: 99%

NeuroPM toolbox: integrating Molecular, Neuroimaging and Clinical data for Characterizing Neuropathological Progression and Individual Therapeutic Needs

Iturria-Medina

Carbonell

Assadi

et al. 2020

Preprint

View full text Add to dashboard Cite

There is a critical need for a better multiscale and multifactorial understanding of neurological disorders, covering from genes to neuroimaging to clinical factors and treatments effects. Here we present NeuroPM-box, a cross-platform, user-friendly and open-access software for characterizing multiscale and multifactorial brain pathological mechanisms and identifying individual therapeutic needs. The implemented methods have been extensively tested and validated in the neurodegenerative context, but there is not restriction in the kind of disorders that can be analyzed. By using advanced analytic modeling of molecular, neuroimaging and/or cognitive/behavioral data, this framework allows multiple applications, including characterization of: (i) the series of sequential states (e.g. transcriptomic, imaging or clinical alterations) covering decades of disease progression, (ii) intra-brain spreading of pathological factors (e.g. amyloid and tau misfolded proteins), (iii) synergistic interactions between multiple brain biological factors (e.g. direct tau effects on vascular and structural properties), and (iv) biologically-defined patients stratification based on therapeutic needs (i.e. optimum treatments for each patient). All models outputs are biologically interpretable. A 4D-viewer allows visualization of spatiotemporal brain (dis)organization. Originally implemented in MATLAB, NeuroPM-box is compiled as standalone application for Windows, Linux and Mac environments: neuropm-lab.com/software. In a regular workstation, it can analyze over 150 subjects per day, reducing the need for using clusters or High-Performance Computing (HPC) for large-scale datasets. This open-access tool for academic researchers may significantly contribute to a better understanding of complex brain processes and to accelerating the implementation of Precision Medicine (PM) in neurology.

show abstract

Gaussian Graphical Models Reveal Inter-Modal and Inter-Regional Conditional Dependencies of Brain Alterations in Alzheimer's Disease

Cited by 35 publications

References 70 publications

Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study

Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study

Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study

NeuroPM toolbox: integrating Molecular, Neuroimaging and Clinical data for Characterizing Neuropathological Progression and Individual Therapeutic Needs

Contact Info

Product

Resources

About