fMRI connectivity, meaning and empiricism: Comments on: Roebroeck et al. The identification of interacting networks in the brain using fMRI: Model selection, causality and deconvolution.
Olivier DavidPublished in: NeuroImage (2011)
Keyphrases
- model selection
- cross validation
- hyperparameters
- brain connectivity
- human brain
- functional magnetic resonance imaging
- activation detection
- bayesian learning
- sample size
- resting state fmri
- brain activity
- machine learning
- parameter estimation
- data analysis
- statistical learning
- brain regions
- mixture model
- error estimation
- regression model
- model selection criteria
- feature selection
- motion segmentation
- generalization error
- functional connectivity
- least squares
- gaussian process
- generalization bounds
- variable selection
- denoising
- selection criterion
- mr images
- causal relationships
- statistical inference
- causal models
- leave one out cross validation
- parameter determination