Analysis tools

This section is being updated by the corresponding editorial team. Please, check back soon.

Several software tools are available for analysis of MEG and other biomagnetic data.

Open Source Software

This section is being updated by the corresponding editorial team. Please, check back soon.


We organize courses for institutions who are willing to develop a user community for Brainstorm. Please contact us to get you started locally.

2012 McGovern Institute, MIT, Boston MA (full day training): Sylvain BailletFrancois Tadel

2011 Montreal Neurological Institute, McGill, QC (full day training): link

Information about the next courses in 2013 (Moscow, Seattle, Florence & Montreal): link



The FieldTrip webpage provides extensive documentation, tutorials and MEG data to run the tutorials. Also a reference documentation and a walkthrough are available on the website.For those interested, the Donders Institutes runs a yearly FieldTrip course, normally mid of April.In addition, there are videos covering basic tutorial teaching on FieldTrip by Julian Keil, Montreal available online.


Human Neocortical Neurosolver (HNN) is a user-friendly software tool that provides a novel solution to this challenge. HNN gives researchers and clinicians the ability to test and develop hypotheses on the circuit mechanism underlying their EEG/MEG data in an easy-to-use environment. The foundation of HNN is a computational neural modelthat simulates the electrical activity of the neocortical cells and circuits that generate the primary electrical currents underlying EEG/MEG recordings.

MEG Processor

MEG Processor is a software package specifically optimized for MEG. Based on native C/C++, it is powerful and productive, with an intuitive, easy-to-use graphic user interface. MEG Processor provides the following functionality: (1)Reads almost all major types of MEG/EEG/MRI/CT data by “drag-drop”; (2) Handles a huge amount of MEG/EEG data (e.g. 1-100 GB); (3) Efficiently localizes brain activities from very low to very high frequency ranges (4) Provides a comprehensive solutions for: multiple filters, averaging, accumulation, time-frequency analysis (e.g. wavelet), virtual sensors, network at both sensor and source level (Graph theory, small world, connectivity), independent component analysis (ICA), volumetric source imaging, distributed source imaging, artificial intelligence (AI), automatic identification of high frequency oscillations (HFOs), ripples, fast ripples and high gamma; (5) Supports parallel computing (e.g. CUDA/video cards) that provide super-fast source scan and “real-time” data analysis; and (6) Provides tools for group comparison, grand-averaging of waveforms and source images, non-parametric multivariate statistics, multivariate pattern analysis and more. For those interested, please follow MEG Processeor youtube channel and contact: to schedule possbile courses from Mecurer LLC.

[1] Xiang et al. Volumetric imaging of brain activity with spatial-frequency decoding of neuromagnetic signals.  J Neurosci Methods. 2015 Jan 15; 239:114-28.


MNE is a software package for processing electrophysiological signals primarily from magnetoencephalographic (MEG) andelectroencephalographic (EEG) recordings [1–2]. It provides a comprehensive solution for data preprocessing, forward modeling (withboundary element models), distributed source imaging, time–frequency analysis, non-parametric multivariate statistics, multivariate patternanalysis, and connectivity estimation. Importantly, this package allows all of these analyses to be applied in both sensor or sourcespace. MNE is developed by an international team, with particular care for computational efficiency, code quality, and readability, as well as the common goal of facilitating reproducibility in neuroscience. The MNE-Python source code is available on github and the documentation is available from the Martinos Center for Biomedical Imaging.

The MNE-CPP is a branch of the MNE family that puts emphasis on real-time processing [3] and the ability to build stand-alone software applications [4].

[1] Gramfort, A and Luessi, M and Larson, E and Engemann, D A and Strohmeier, D and Brodbeck, C and Parkkonen, L and Hämäläinen, M S (2014), MNE software for processing MEG and EEG data, NeuroImage, vol. 86, pp. 446-460.

[2] Gramfort, A and Luessi, M and Larson, E and Engemann, D A and Strohmeier, D and Brodbeck, C and Goj, R and Jas, M and Brooks, T and Parkkonen, L and Hämäläinen, M S (2013), MEG and EEG data analysis with MNE-­Python, Frontiers in Neuroscience, vol. 7, no. 267.

[3] Dinh C, Esch L, Rühle J, Bollmann S, Güllmar D, Baumgarten D, Hämäläinen MS, Haueisen J: Real-Time Clustered Multiple Signal Classification (RTC-MUSIC). Brain Topography, DOI 10.1007/s10548-017-0586-7, 2017 

[4] Esch L, Sun L, Klüber V, Lew S, Baumgarten D, Grant PE, Okada Y, Haueisen J, Hämäläinen MS, Dinh C: MNE Scan: Software for Real-Time Processing of Electrophysiological Data. Journal of Neuroscience Methods, 303:55–67, 2018


OpenMEEG is a software package for solving forward problems in MEG, EEG, ECoG, intracerebral EEG, and more generally, quasistatic electromagnetics. It uses a boundary element model, assuming  piecewise constant conductivity, and it can accommodate non-nested and zero-conductivity regions. OpenMEEG is able to compute forward problems with high accuracy, due to the use of the symmetric BEM and advanced numerical methods [1-2]. Several toolboxes, for instance BrainStorm and Fieldtrip, have integrated OpenMEEG as forward solver. The OpenMEEG source code (C++) is freely accessible on github and precompiled binaries are available too.

[1] Kybic J., Clerc M., Abboud T., Faugeras O., Keriven R., Papadopoulo T. A common formalism for the integral formulations of the forward EEG problem. IEEE Transactions on Medical Imaging, 24:12-28, 2005. 

[2] Gramfort A., Papadopoulo T., Olivi E., Clerc M. OpenMEEG: opensource software for quasistatic bioelectromagnetics, BioMedical Engineering OnLine 45:9, 2010.


SPM for MEG analysis

The Matlab analysis toolbox SPM (Statistical Parametric Mapping) - initially designed for the analysis of MRI data - was expanded to cover the analysis of MEG/EEG data. The Wellcome Trust Centre for Neuroimaging at University College London incorporating the Functional Imaging Laboratory (FIL) offers a specific yearly SPM course on the analyses of MEG and EEG data.

2012 Slides and videos of the course in London, May 2012 are available online. The faculty included e.g. Vladimir Litvak, Gareth Barnes, Will Penny, Christophe Philipps, and Jason Tyler. Slides from a MEEG course in Lyon in April can be found here

2011 Two courses have been offered (i) in Brussels featuring Vladimir Litvak, Jéréremie Matteau, Jason Taylor and Christophe Philipps and (ii) one in London where the faculty included Vladimir Litvak, Gareth Barnes, Rik Henson, Will Penny, Stefan Kiebel, Jeremie Matteau, Jean  Daunizeau, Guillaume Flandin, Christophe Philipps.

2010 The faculty of the course in May 2010 includes e.g. Vladimir Litvak, Gareth Barnes, Rik Henson, Will Penny, Stefan Kiebel, Jérémie Matteau, Jean  Daunizeau, Guillaume Flandin, Christophe Philipps and slides of the course can be found here.

2009 To see slides from 2009, click here.

Commercial Software

This section is being updated by the corresponding editorial team. Please, check back soon.

Below is fragmented list of commercial software solutions for electrical and magnetic source analysis using MEG and EEG. The main features of commercial software beyond the expected quality and robustness of its workflow are documentation, customer support and in some cases, compliance with clinical standards, as those imposed by the FDA in the USA.

This is obviously an incomplete list of applications. If you are representing a software vendor and would like to be listed here, please contact us.

Test datasets

This section is being updated by the corresponding editorial team. Please, check back soon.

  • MEG SIM. Web-portal resource that is available for testing various algorithms or methods for source localization, time-course characterization, using very realistic simulated data (i.e., based on individual MRIs and averaged spontaneous MEG data).
  • MMN dataset used for tutorial/publications applying DCM for ERP analysis using SPM
  • Kymata Datasets used to test hypotheses in the Kymata atlas. Comprising of (averaged) EEG and MEG sensor data, and current density reconstructions. The participants are healthy human adults listening to the radio and/or watching films.
  • Human Connectome Project Datasets  Over a 3-year span (2012-2015), the Human Connectome Project (HCP) scanned 1,200 healthy adult subjects. The available data includes MR structural scans, behavioural data and (on a subset of the data) resting state and/or task MEG data.