A simple method for calculating the depth of EEG sources using minimum norm estimates (MNE).

Area of Science:

• Neuroscience
• Biophysics
• Signal Processing

Background:

• Neural source localization commonly uses dipolar or minimum norm methods.
• Dipolar models require prior source number information and are best for small cortical areas.
• Minimum norm techniques struggle with uncorrelated noise and tend to surface-localize sources.

Purpose of the Study:

• To develop a novel mathematical procedure for estimating the depth of neural sources.
• To overcome limitations of existing electroencephalographic (EEG) source localization techniques.
• To improve the accuracy of identifying the true 3D location of brain activity generators.

Main Methods:

• A mathematical procedure based on the dispersion of minimum norm solutions was developed.
• Successively deeper surfaces were used for localization, avoiding traditional regularization matrices.
• The method was evaluated using simulated single and double dipolar sources in spherical and realistic head models.

Main Results:

• The new technique successfully estimated source depth with a mean accuracy of approximately 10 mm.
• Accurate localization was achieved even in disadvantageous simulated scenarios.
• The method demonstrated effectiveness across different head models (spherical and realistic).

Conclusions:

• The proposed method offers a promising approach for accurately estimating the depth of neural generators.
• This technique addresses key limitations of current EEG source localization methods.
• Improved depth estimation can enhance the understanding of brain activity and neural processes.