MatlabElectrodeSelection

The MatlabElectrodeSelection Component enables semi to fully automated electrode localization implemented through matlab.

Automated Electrode Localization

After an initial threshold of the CT, we create 3D centroids by finding connected voxels.

Every Centroid identified after thresholding is used as a seed to create a connected graph based on some conditions:

For every centroid, we do the following: For Grids: Find 2 other centroids that create a 90-degree angle with the starting point. The search radius is the inter-electrode distance. For each of the valid points we find we look for other points fitting the criteria - this will create a connected graph. We can then traverse the graph to find out how many points are part of the grid. In addition, there is an uncertainty parameter: which is set to 0.15 or something close to it. This just means we allow up to 15% error in the measurements (since we are not taking into account the curvature of grids, that allows us to still get connected parts that are somewhat curved).

For Strips and SEEG electrodes, we just use 180 degrees instead of 90.

And lastly, we will sweep through different threshold values of the CT to find where we get the best results. The Cost function for that optimization searches for the minimum of the difference in the number of electrodes found versus the known number of electrodes.

Dependencies

• Image Processing Toolbox
• Statistics and Machine Learning Toolbox
• Computer Vision Toolbox

Inputs and Outputs

• Inputs: Post-OP CT, ElectrodeDefinition

• OptionalInputs: Surface, Trajectory, Existing electrode Locations

• Outputs: Electrode locations

Settings

ElectrodeDefinitionIdentifier

Default: ElectrodeDefinition

CTIdentifier

Post-OP CTs

Default: CT

SurfaceIdentifier

Optional identifier surface. This helps to orient yourself during manual selection

Default: Surface

ElectrodeLocationIdentifier

Optional identifier which uses existing electrode locations Default: ElectrodeLocation

TrajectoryIdentifier

Optional Identifier for trajectories (for example rosa trajectories). This allows fully automated localization of electrodes.

Default: Trajectory

XML Example

<Component Type="MatlabElectrodeSelection">
  <Name>"Localize Electrodes"</Name>

</Component>