Skip to content

Package: source_localization

EEG source localization algorithms and related nodes.

These are domain-specific nodes that are designed to solve the inverse problem of estimating the source activity given some EEG (or MEG) observations. Also includes some nodes that act on the resulting solutions (for instance, operate on regions of interest). Nodes in this category typically process time series that are consequently expected to have a time axis and a space axis. The space axis can variably index sensor channels, regions of interest, or vertices/voxels in source space, depending on the context. This category also includes some nodes for managing location information such as the one-stop-shot AssignChannelLocations node. Nodes that expect head models expect these models in a NeuroPype-specific file format (.hm) that can be interpreted as a MATLAB .mat file for customization. Predefined head models are available in the resources/headmodels folder.

Nodes in this package:

  • Assign Channel Locations
    Assign standard channel locations from a pre-defined sensor montage file.

  • ComputeRegions (Meta)
    A meta-node that performs source localization for given regions of interest using sLoreta, and optionally groups them based on a given mapping.

  • Coregister Existing Locations
    Co-register the locations in the given data with the NeuroPype internal coordinate system.

  • eLORETA
    Estimate brain source activity using the Exact Low-Resolution Brain Electromagnetic Tomography (eLORETA) algorithm.

  • LCMV Beamforming
    Estimate brain source activity using the linearly-constrained minimum variance (LCMV) beamforming algorithm.

  • Project Spatial Locations Onto Headmodel Mesh
    Projects each position on the space axis onto its nearest vertex on the mesh for the given headmodel.

  • ROI Activations
    Compute the activations of regions of interest (ROIs) in the given atlas from current source density.

  • ROIs To Vertices
    Expand ROI activations such that each ROI gets broadcast to all of its vertices.

  • Remove Unlocalized Channels
    Remove channels from the data that have no location assigned to them.

  • sLORETA
    Estimate brain source activity using the Standardized Low-Resolution Brain Electromagnetic Tomography (sLORETA) algorithm.

  • Select Regions of Interest
    Pick a subset of regions of interest in the given data.

  • Select ROI Groups
    Selects among major groups of regions of interest taken from the Desikan-Killiany atlas.