Mne Library (`MNELIB`)

The Mne library contains the core neuroscience data structures used throughout MNE-CPP — source spaces, source estimates, forward solutions, inverse operators, hemispheres, BEM surfaces, covariance matrices, epochs, projections, and many supporting geometry types. It is the C++ counterpart of the structures defined in the original MNE-C libmne and the core mne Python package. All classes reside in the MNELIB namespace.

Architecture

The library mirrors the data model of the original MNE software closely. Most classes are direct C++ translations of the MNE-C structs with added Qt/Eigen integration.

Class Inventory

Source Spaces

Class	Description	MNE-Python	MNE-C
`MNESourceSpace`	Single surface or volume source space with vertex positions, normals, triangles, and neighbour lists	`SourceSpaces[0]`	`mneSurfaceOrVolume`
`MNESourceSpaces`	Container for multiple source spaces (typically left + right hemisphere)	`mne.SourceSpaces`	Multiple `mneSurfaceOrVolume`
`MNEHemisphere`	Cortical hemisphere source space with patch clustering and decimated vertex sets	Via `SourceSpaces`	`mneSurfaceOrVolume`
`MNESurfaceOrVolume`	Base class for any surface or volume mesh with common geometry fields	Base of `SourceSpaces`	`mneSurfaceOrVolume`

Surfaces & Geometry

Class	Description	MNE-Python	MNE-C
`MNESurface`	Triangulated surface mesh (vertices, normals, triangles)	`read_surface()`	`mneSurface`
`MNETriangle`	Per-triangle data: edge vectors, face normal, area, centroid	Internal	`mneTriangle`
`MNEVolGeom`	Volume geometry: voxel-to-RAS transforms, dimensions, voxel size	`VolSourceEstimate` metadata	`mneVolGeom`
`MNEPatchInfo`	Cortical patch: member vertices, area, average normal deviation	`Label` (partial)	`mnePatchInfo`
`MNENearest`	Nearest-neighbour record linking original vertices to decimated source points	Adjacency matrices	`mneNearest`
`MNEClusterInfo`	Cluster assignment with labels and statistics	`mne.stats.cluster_level` results	`mneClusterInfo`

BEM Surfaces

Class	Description	MNE-Python	MNE-C
`MNEBem`	BEM model container holding multiple tissue surfaces	`mne.read_bem_solution()`	Multiple `mneBemSurface`
`MNEBemSurface`	Single BEM surface (scalp, outer skull, inner skull) with conductivity and triangle metadata	BEM surface in solution	`mneBemSurface`

Forward & Inverse Operators

Class	Description	MNE-Python	MNE-C
`MNEForwardSolution`	Lead-field matrix with geometry, channel info, and source space reference	`mne.Forward` / `mne.read_forward_solution()`	`mneForwardSolution`
`MNEInverseOperator`	SVD-decomposed whitened lead field with source and noise covariance	`mne.minimum_norm.InverseOperator`	`mneInverseOperator`

Raw Data & Measurement

Class	Description	MNE-Python	MNE-C
`MNERawData`	Raw continuous data with filtering, projection, and CTF/SSS compensation	`mne.io.Raw`	`mneRawData`
`MNERawInfo`	Metadata about raw data (sampling rate, channels, transforms)	`Raw.info`	`mneRawInfo`
`MNERawBufDef`	Buffer segment definition for memory-mapped raw data access	Internal	`mneRawBufDef`
`MNEMeasData`	Measurement data container with associated inverse data	`mne.Evoked` / `mne.Epochs`	`mneMeasData`
`MNEMeasDataSet`	Single measurement epoch with data matrix and metadata	Per-epoch data	`mneMeasDataSet`
`MNEEpochData`	Single trial epoch with time range and artifact rejection status	Item in `mne.Epochs`	`mneEpochData`
`MNEEpochDataList`	List of epochs with rejection thresholds, baseline, and I/O routines	`mne.Epochs`	Multiple `mneEpochData`
`MNEMneData`	Computed MNE inverse data: projection, prediction, SNR, regularisation parameters	Attributes in `SourceEstimate`	`mneMneData`

Morph Maps & Surface Operations

Class	Description	MNE-Python	MNE-C
`MNEMorphMap`	Vertex-to-vertex mapping between two FreeSurfer surfaces for morphing source estimates	`mne.read_morph_map()`	`morphMapRec`
`MNESurfacePatch`	Patch linking vertices to a source point on cortical surface	Embedded in `SourceSpaces`	Part of `mnePatchInfo`
`MNEProjectToSurface`	Projection of arbitrary 3D points onto a triangulated surface mesh	`_project_onto_surface()`	`mne_project_to_surface`
`MNEIcp`	Iterative Closest Point algorithm for point-cloud-to-surface registration (co-registration)	ICP via `mne.coreg`	—

Projection & Compensation

Class	Description	MNE-Python	MNE-C
`MNEProjOp`	SSP projection operator managing multiple projection items	`mne.read_proj()`	`mneLinspace`
`MNEProjItem`	Single SSP vector with kind, description, and active flag	One SSP component	`mneLinspaceRec`
`MNEProjData`	Precomputed auxiliary projection data for efficient repeated application	Internal	—
`MNECovMatrix`	Noise or source covariance with eigendecomposition, channel sub-selection, and whitening	`mne.Covariance`	`mneCovMatrix`
`MNECTFCompData`	Single CTF compensation element (matrix + metadata)	`mne.Info['comps'][i]`	`mneCTFCompData`
`MNECTFCompDataSet`	Collection of CTF compensation matrices	`apply_gradient_compensation()`	Multiple `mneCTFCompData`
`MNESssData`	Maxwell filtering (SSS) data for gradiometer/magnetometer reduction	SSS via `maxwell_filter()`	`mneSssData`

Derivations & Channels

Class	Description	MNE-Python	MNE-C
`MNEDeriv`	Single virtual channel derivation with sparse coefficient vector	Custom channels	`mneDeriv`
`MNEDerivSet`	Collection of channel derivations	—	Multiple `mneDeriv`
`MNEChSelection`	Channel selection with regex pattern and picked indices	`pick_channels()`	`mneChSelection`

Named Matrices & Vectors

Class	Description	MNE-Python	MNE-C
`MNENamedMatrix`	Dense matrix with labelled row and column names	Internal	`mneNamedMatrix`
`MNENamedVector`	Vector with element names	Internal	`mneNamedVector`
`MNESparseNamedMatrix`	Sparse matrix with labelled rows/columns	Internal	`mneSparseNamedMatrix`

Visualisation Helpers

Class	Description	MNE-Python	MNE-C
`MNEMshDisplaySurface`	Surface mesh for 3D rendering with colour mapping and morphing	`Brain` / `add_surface()`	`mshDisplaySurface`
`MNEMshDisplaySurfaceSet`	Collection of display surfaces for multi-hemisphere visualisation	Brain surfaces	Multiple `mshDisplaySurface`
`MNEMshLight` / `MNEMshLightSet`	Light sources for 3D scene	Lighting config	`mshLight`
`MNEMshEyes`	Camera/eye position for 3D rendering	Camera viewpoint	`mshEyes`
`MNEMshPicked`	Picked vertex on displayed surface (index, coordinates, value)	Click selection	`mshPicked`
`MNEMshColorScaleDef`	Colour scale definition with value range and colour mapping	Colormap	`mshColorScaleDef`
`MNELayout` / `MNELayoutPort`	2D channel layout for topographic displays	`mne.channels.read_layout()`	`mneLayout`

Events & Triggers

Class	Description	MNE-Python	MNE-C
`MNEEvent`	Single trigger/stimulus event marker with sample, before, and after values	Event tuple	`mneEvent`
`MNEEventList`	Ordered collection of event markers	`mne.read_events()` results	Multiple `mneEvent`

File I/O & Utilities

Class	Description	MNE-Python	MNE-C
`MNEDescriptionParser`	Parser for MNE-C style `.ave` and `.cov` description files	—	`interpret.c`
`MNEProcessDescription`	Parsed averaging/covariance process description data	—	—
`MNEFilterDef`	Filter definition (highpass, lowpass, notch parameters)	Filter specs	`mneFilterDef`

Clustering

Class	Description	MNE-Python	MNE-C
`RegionData` / `RegionMT`	Input parameters for KMeans-based cortical region clustering	`mne.SourceSpaces` clustering	—
`RegionDataOut` / `RegionMTOut`	Output of KMeans-based source clustering on cortical regions	Cluster results	—

Static Utility Class

Class	Description
`MNE`	Static utility wrapper providing MATLAB-toolbox-like functions: `combine_xyz()`, `find_source_space_hemi()`, `make_compensator()`, `get_current_comp()`, `make_block_diag()`, `read_events()`, `setup_volume_source_space()`

Usage Example

#include <mne/mne.h>

using namespace MNELIB;
using namespace FIFFLIB;

// Read source spaces
MNESourceSpaces srcSpaces;
MNE::read_source_spaces("sample-src.fif", srcSpaces);

qDebug() << "Hemispheres:" << srcSpaces.size();
qDebug() << "LH vertices:" << srcSpaces[0].nuse;
qDebug() << "RH vertices:" << srcSpaces[1].nuse;

// Read forward solution
MNEForwardSolution fwd =
    MNEForwardSolution::read("sample_audvis-meg-eeg-oct-6-fwd.fif");

// Read inverse operator
MNEInverseOperator invOp =
    MNEInverseOperator::read("sample_audvis-meg-eeg-oct-6-inv.fif");

// Read epochs
FiffRawData raw("sample_audvis_raw.fif");
Eigen::MatrixXi events;
MNE::read_events("sample_audvis_raw-eve.fif", events);

MNEEpochDataList epochs;
epochs = MNEEpochDataList::readEpochs(raw, events,
                                       -0.2, 0.5, /*eventType=*/1);

Doxygen Reference

For method signatures, inheritance diagrams, and source-level documentation see the auto-generated MNELIB namespace in the Doxygen API reference.

Architecture​

Class Inventory​

Source Spaces​

Surfaces & Geometry​

BEM Surfaces​

Forward & Inverse Operators​

Raw Data & Measurement​

Morph Maps & Surface Operations​

Projection & Compensation​

Derivations & Channels​

Named Matrices & Vectors​

Visualisation Helpers​

Events & Triggers​

File I/O & Utilities​

Clustering​

Static Utility Class​

Usage Example​

Doxygen Reference​

See Also​