The SESAMEEG package: a probabilistic tool for source localization and uncertainty quantification in M/EEG

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Neuroscience

Background:

• Magnetoencephalography (M/EEG) data analysis relies heavily on accurate source localization.
• Clinical applications, such as epilepsy pre-surgical evaluation, demand robust source localization methods.
• Existing methods may have limitations in accuracy, parameter sensitivity, or uncertainty quantification.

Purpose of the Study:

• To introduce and describe the SESAME (SEquential SemiAnalytic Montecarlo Estimator) algorithm for M/EEG source localization.
• To detail the comprehensive output of the SESAME algorithm and its practical applications.
• To provide a user-friendly guide to SESAME implementation and interpretation.

Main Methods:

• Bayesian source localization using the SESAME algorithm.
• Sequential SemiAnalytic Montecarlo estimation.
• Frequency domain analysis for neural oscillations.

Main Results:

• SESAME demonstrates high accuracy in localizing focal neural sources.
• The method quantifies the uncertainty associated with source reconstruction.
• SESAME is robust to input parameter variations and accepts user-defined search regions.
• The algorithm can identify generators of neural oscillations in the frequency domain.

Conclusions:

• SESAME offers a powerful and accurate Bayesian approach to M/EEG source localization.
• Its flexibility, uncertainty quantification, and open-source availability (SESAMEEG) make it valuable for research and clinical applications.
• A thorough understanding of SESAME's output is crucial for effective utilization in M/EEG analysis pipelines.