User's Guide‎ > ‎Analysis‎ > ‎

3 - Interpolation

Interpolation (necessary after rejection of complete EEG channels) 

-          Tools > Interpolating EEG files

-          XYZ-file: insert a cap with your appropriate no. of electrodes (e.g. BIOSEMI128new.xyz)

-          Define the electrodes you want to interpolate (usually includes electrodes excluded from the 'test to threshold' automatic rejection applied during the averaging process) 

-          Choose Orientation as in file or using landmarks (useful if you have a different 'from' and 'to' space)

-          Choose To-Space “Identical to 'From' space” * (if you don’t want to downsample your electrode setup)

-          Choose 3D spline interpolation (if not indicated otherwise)

-          Output files get an automatic infix like *.i3DS2.To160.eph or *.i3DS2.To128.eph

 

TIPS: 

  • Electrodes with large SE, abrupt modulations, or incongruent with surrounding electrodes should generally be interpolated. If a lot of triggers were rejected during averaging, you should consider re-averaging with these electrodes excluded. 
  • Linking your .eph with a cap (using a LinkMany file) may help to detect "bad" electrodes as for example flat or bridged electrodes pop out in map view.
  • Electrodes at the border of the electrode array should be interpolated with caution since neighbor electrodes used for the interpolation are missing -and in this case it's more an extrapolation than an interpolation: signal is 'created'
  • You should avoid interpolating more than 10% of the total number of electrodes

 

 * "Identical to 'From' space" means that the same xyz-coordinates file is used for the interpolation. If the 'To' and the 'From' space are identical, you can chose 'As in File' for the 'Orientation, Center & Normalization' parameter and usually no need to apply a co-registration.  If you want to downsample directly to a smaller number of electrodes you can add another .xyz file  (i.e. without unwanted electrodes). 

 


Comments