Limits...
Bayesian networks in neuroscience: a survey.

Bielza C, LarraƱaga P - Front Comput Neurosci (2014)

Bottom Line: Thanks to their generality, Bayesian networks can accommodate continuous and discrete variables, as well as temporal processes.Also, we examine how a user can take advantage of these networks for reasoning by exact or approximate inference algorithms that propagate the given evidence through the graphical structure.Here we survey key research in neuroscience where Bayesian networks have been used with different aims: discover associations between variables, perform probabilistic reasoning over the model, and classify new observations with and without supervision.

View Article: PubMed Central - PubMed

Affiliation: Departamento de Inteligencia Artificial, Universidad Politecnica de Madrid Madrid, Spain.

ABSTRACT
Bayesian networks are a type of probabilistic graphical models lie at the intersection between statistics and machine learning. They have been shown to be powerful tools to encode dependence relationships among the variables of a domain under uncertainty. Thanks to their generality, Bayesian networks can accommodate continuous and discrete variables, as well as temporal processes. In this paper we review Bayesian networks and how they can be learned automatically from data by means of structure learning algorithms. Also, we examine how a user can take advantage of these networks for reasoning by exact or approximate inference algorithms that propagate the given evidence through the graphical structure. Despite their applicability in many fields, they have been little used in neuroscience, where they have focused on specific problems, like functional connectivity analysis from neuroimaging data. Here we survey key research in neuroscience where Bayesian networks have been used with different aims: discover associations between variables, perform probabilistic reasoning over the model, and classify new observations with and without supervision. The networks are learned from data of any kind-morphological, electrophysiological, -omics and neuroimaging-, thereby broadening the scope-molecular, cellular, structural, functional, cognitive and medical- of the brain aspects to be studied.

No MeSH data available.


Related in: MedlinePlus

Example of HMM. The real state of three ROIs is unknown (gray nodes) and we indirectly observe them with BOLD responses in an fMRI experiment (white nodes).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4199264&req=5

Figure 3: Example of HMM. The real state of three ROIs is unknown (gray nodes) and we indirectly observe them with BOLD responses in an fMRI experiment (white nodes).

Mentions: Example. Figure 3 shows an example of HMM. The model represents a simple functional connectivity analysis, where 3 ROIs have been identified. Gray nodes are hidden variables and represent the unknown real neural activity of a ROI, e.g., whether the region is activated or not. White nodes are the observed measures Oi of each region i, e.g., the BOLD response in fMRI experiments. This simple model (parallel HMM) factorizes the state space into multiple independent temporal processes without connections in-between. Other versions of HMMs are more complex.


Bayesian networks in neuroscience: a survey.

Bielza C, LarraƱaga P - Front Comput Neurosci (2014)

Example of HMM. The real state of three ROIs is unknown (gray nodes) and we indirectly observe them with BOLD responses in an fMRI experiment (white nodes).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4199264&req=5

Figure 3: Example of HMM. The real state of three ROIs is unknown (gray nodes) and we indirectly observe them with BOLD responses in an fMRI experiment (white nodes).
Mentions: Example. Figure 3 shows an example of HMM. The model represents a simple functional connectivity analysis, where 3 ROIs have been identified. Gray nodes are hidden variables and represent the unknown real neural activity of a ROI, e.g., whether the region is activated or not. White nodes are the observed measures Oi of each region i, e.g., the BOLD response in fMRI experiments. This simple model (parallel HMM) factorizes the state space into multiple independent temporal processes without connections in-between. Other versions of HMMs are more complex.

Bottom Line: Thanks to their generality, Bayesian networks can accommodate continuous and discrete variables, as well as temporal processes.Also, we examine how a user can take advantage of these networks for reasoning by exact or approximate inference algorithms that propagate the given evidence through the graphical structure.Here we survey key research in neuroscience where Bayesian networks have been used with different aims: discover associations between variables, perform probabilistic reasoning over the model, and classify new observations with and without supervision.

View Article: PubMed Central - PubMed

Affiliation: Departamento de Inteligencia Artificial, Universidad Politecnica de Madrid Madrid, Spain.

ABSTRACT
Bayesian networks are a type of probabilistic graphical models lie at the intersection between statistics and machine learning. They have been shown to be powerful tools to encode dependence relationships among the variables of a domain under uncertainty. Thanks to their generality, Bayesian networks can accommodate continuous and discrete variables, as well as temporal processes. In this paper we review Bayesian networks and how they can be learned automatically from data by means of structure learning algorithms. Also, we examine how a user can take advantage of these networks for reasoning by exact or approximate inference algorithms that propagate the given evidence through the graphical structure. Despite their applicability in many fields, they have been little used in neuroscience, where they have focused on specific problems, like functional connectivity analysis from neuroimaging data. Here we survey key research in neuroscience where Bayesian networks have been used with different aims: discover associations between variables, perform probabilistic reasoning over the model, and classify new observations with and without supervision. The networks are learned from data of any kind-morphological, electrophysiological, -omics and neuroimaging-, thereby broadening the scope-molecular, cellular, structural, functional, cognitive and medical- of the brain aspects to be studied.

No MeSH data available.


Related in: MedlinePlus