As it relates to neuroscience, this code base aims to perform the following general tasks:
- Function Connectivity Mapping
- looking for network-based relationships among variables of interest. This is known in the more general case as "Association Discovery".
- Feature Selection
- choosing the appropriate features (variables) to use in classifying a target variable, or otherwise reducing the dimensionality of a large dataset.
- Supervised Classification
- predict class labels of new data from training data.
- Unsupervised Classification
- learning from data where the ground truth is unknown. One example of this task is "clustering" - assigning data into clusters of similar observations.
- Inference
- answering marginal or conditional probability queries, such as "what is the probability a person will develop dimentia given that he/she is 60 years old and had a stroke within the last five years?"
The goal is to achieve full availability of state-of-the-art Bayesian network functionality for seamless use with all types of neuroscience data.
- Morphological Data
- Electrophysiological Data
- Genomics/Proteomics/Transcriptomics Data
- Neuroimaging Data
- fMRI
- MRI
- EEG
- Others
- ECoG, EMG, JME, TCD, DTI, PET