MNE-CPP Manual

Overview

MNE-CPP provides a comprehensive set of tools for preprocessing and averaging of MEG and EEG data and for constructing cortically-constrained minimum-norm estimates. The software is based on anatomical MRI processing, forward modelling, and source estimation methods. The anatomical MRI processing routines depend on tools provided by the FreeSurfer software.

This manual gives an overview of the software modules included with MNE-CPP, describes a typical workflow for a novice user, and provides detailed information about the individual software modules.

Heritage

MNE-CPP is a C++ port of the original MNE software suite created by Matti Hämäläinen at the Martinos Center for Biomedical Imaging. This manual is adapted from the original MNE Manual and the MNE-Python documentation, preserving the valuable background information and mathematical descriptions while documenting the C++ implementations available in MNE-CPP.

Software Components

The MNE-CPP software suite includes the following main components:

GUI Applications

Application	Description
mne_scan	Real-time acquisition and processing of MEG/EEG data with plugin-based architecture
mne_analyze	Sensor- and source-level analysis of pre-recorded MEG/EEG data (under development)
mne_inspect	3D brain visualization and source analysis
mne_browse	Browsing and visualization of raw MEG/EEG FIFF data files (under development)

Command-Line Tools

See the Command-Line Tools Overview for full details. The tools are organized by domain:

Forward Modeling

Tool	Description
mne_watershed_bem	Creates BEM surfaces using FreeSurfer's watershed algorithm
mne_flash_bem	Creates BEM surfaces using multi-echo FLASH MRI sequences
mne_surf2bem	Converts FreeSurfer surfaces to BEM FIFF files
mne_setup_forward_model	Sets up the BEM for forward modeling
mne_setup_mri	Sets up FreeSurfer MRI data for MNE processing
mne_forward_solution	Computes the MEG/EEG forward solution
mne_make_source_space	Creates a source space for forward/inverse computations
mne_make_sphere_bem	Creates a spherically symmetric BEM model
mne_prepare_bem_model	Prepares a BEM model for forward computation
mne_average_forward_solutions	Averages forward solutions
mne_check_surface	Checks BEM surface integrity

Inverse

Tool	Description
mne_inverse_operator	Assembles the inverse operator
mne_compute_mne	Computes MNE/dSPM/sLORETA from evoked data
mne_compute_raw_inverse	Computes inverse solutions from raw or evoked data
mne_dipole_fit	Performs electric current dipole fitting
mne_sensitivity_map	Computes sensitivity maps for MEG/EEG
mne_smooth	Smooths source estimates on the cortical surface
mne_volume_data2mri	Converts volumetric source estimates to MRI coordinates

Preprocessing

Tool	Description
mne_anonymize	Removes or modifies personal health information from FIFF files
mne_process_raw	Filters, averages, and processes raw MEG/EEG data
mne_mark_bad_channels	Marks bad channels in a FIFF file
mne_compensate_data	Applies or removes software gradient compensation
mne_cov2proj	Converts a noise covariance matrix to a projection operator
mne_add_to_meas_info	Adds information to a FIFF measurement file
mne_create_comp_data	Creates compensation data for CTF systems
mne_fix_mag_coil_types	Fixes magnetometer coil type entries
mne_insert_4D_comp	Inserts 4D compensation data into a FIFF file
mne_rename_channels	Renames channels in a FIFF file

Conversion

Tool	Description
mne_edf2fiff	Converts EDF (European Data Format) files to FIFF
mne_brain_vision2fiff	Converts Brain Vision files to FIFF
mne_ctf2fiff	Converts CTF data to FIFF
mne_ctf_dig2fiff	Converts CTF digitization data to FIFF
mne_kit2fiff	Converts KIT/Yokogawa data to FIFF
mne_raw2mat	Converts raw FIFF data to MATLAB format
mne_eximia2fiff	Converts Nexstim eXimia data to FIFF
mne_tufts2fiff	Converts Tufts University data to FIFF
mne_convert_dig_data	Converts digitizer data between formats
mne_convert_surface	Converts surface files between formats
mne_make_cor_set	Creates a set of COR files from MRI data

Surface

Tool	Description
mne_annot2labels	Converts FreeSurfer annotation files to label files
mne_add_patch_info	Adds patch information to a source space
mne_volume_source_space	Creates a volumetric source space
mne_morph_labels	Morphs label files between subjects
mne_make_morph_maps	Creates morphing maps between subjects
mne_make_eeg_layout	Creates EEG channel layout files

Info

Tool	Description
mne_show_fiff	Lists the contents of a FIFF file
mne_list_source_space	Lists source space information
mne_compare_fif_files	Compares two FIFF files
mne_collect_transforms	Collects coordinate transforms from a FIFF file
mne_list_bem	Lists BEM surface information

Server & Simulation

Tool	Description
mne_rt_server	Real-time FIFF data streaming server
mne_simu	Simulates MEG/EEG data

Prerequisites

To make full use of the MNE-CPP software, you will need:

• FreeSurfer: Required for anatomical MRI processing, cortical surface reconstruction, and the creation of BEM meshes. Available at https://surfer.nmr.mgh.harvard.edu/.

• MEG/EEG Data: Data in FIFF format (the native format of Neuromag/Elekta/MEGIN MEG systems). Data from other systems can be converted using appropriate tools (e.g., mne_edf2fiff for EDF data).

• Environment Variables: Several MNE-CPP tools rely on the following environment variables:

  • SUBJECTS_DIR — The directory containing FreeSurfer subject reconstructions
  • SUBJECT — The name of the current subject
  • FREESURFER_HOME — The FreeSurfer installation directory

Chapter Overview

• The Typical MEG/EEG Workflow — A step-by-step guide through a typical analysis pipeline
• The Forward Solution — Coordinate systems, BEM models, and forward computation
• The Minimum-Norm Estimates — Mathematical background of inverse estimation methods
• Signal-Space Projection — The SSP method for noise rejection
• Command-Line Tools Reference — Detailed reference for all command-line tools
• MNE Browse Raw — The raw data browser application
• MNE Analyze — The analysis and visualization application

Acknowledgments

The development of the MNE software has been supported by the National Center for Research Resources (NCRR), the NIH (grants R01EB009048, R01EB006385, R01HD40712), and the U.S. Department of Energy. We are grateful to the MNE software users at the Martinos Center and other institutions for their collaboration and valuable feedback on the software and its documentation.