This section summarizes the MATLAB commands and data structures used for each stage of head modeling using the NFT toolbox. The function reference can be found in Appendix B

The toolbox functions have names all lowercase, words separated by underscore. The GUI function names start with capital letters. The user interface functions all begin with the name of the module, such as bem_, mesh_, segm_, warping_ while utility functions are prefixed by util, for instance BEM utility functions start with utilbem_. All functions have help sections describing the usage, inputs and outputs, compatible with the help2html() conventions. They also include a license block.

In user interface functions (bem_, etc.) the input arguments are validated before use. The utility functions, which are mostly used internally do minimal validation.

The main user interface for the segmentation is initiated using Segmentation() command which opens up the GUI for segmentation.

While most of the BEM matrix generation functionality can be initiated from the GUI interface, each operation can also be performed through matlab functions. After loading the MR image the following functions are called respectively: segm_aniso_filtering(), segm_scalp(), segm_brain(), segm_outer_skull(), segm_inner_skull(), segm_final_skull().

The main user interface for mesh generation is initiated using Mesh_generation() command which opens up the GUI for mesh generation. The function loads the segmentation and generates meshes for scalp, skull, CSF and brain. If the user wants to refine the meshes locally, mesh_local_refinement() function is called. The topology of the generated meshes are checked by mesh_final_correction(), and finally, the total head mesh is written in the format described in A using the function mesh_read_write().

The main user interface for the co-registration of electrode locations with MR images is initiated using Coregistration() command which opens up the GUI for co-registration.

The main user interface for the warping of a template head model is initiated using Warping_mesh() command which opens up the GUI for warping functions. After loading the electrode locations the MNI mesh is loaded. Using warping_main_function() function the warping parameters and the warped mesh are calculated.

The main user interface for the forward model generation is initiated using Forward_Problem_Solution() command. While most of the BEM matrix generation functionality can be initiated from the GUI interface, each operation also be performed through matlab functions.

The functions used by the BEM module are used for creating the BEM structures, running the solver to generate the model matrices, and solving for single or multiple dipoles.

The state of the forward solution is stored in the structures, and no global variables are used by the toolbox. Since the contents of the structure arguments may change during a function call, most interface functions return the structure so that the changes can be preserved (MATLAB has no OUT arguments).

A set of mesh files can be loaded with the bem_load_mesh() function which returns a mesh structure.

The model structure which combines the mesh, conductivities and solver IPA parameters is obtained using the bem_create_model() function. Once the model structure is obtained, it is possible to invoke the solver using the bem_generate_eeg_matrices() function to generate the BEM matrices. Individual matrices can be loaded into the model structure using the bem_load_model_matrix() function.

The session structure is used for solving the forward problem at a given set of sensors. The structure is created using bem_create_session() from a model and a list of sensors. The list of sensors can be generated from a list of nodes using the bem_smatrix_from_nodes() function.

The transfer matrix for the sensors specified in the session can be generated using the bem_generate_eeg_transfer_matrix() function which computes and saves the transfer matrix. The computed transfer matrix can be loaded into the session structure using the bem_load_transfer_matrix() function.

There are two functions for obtaining forward solutions. The bem_solve_dipoles_eeg() function computes the sensor potentials due to the activation of a number of dipoles. The bem_solve_lfm_eeg() function is suitable for generating a Lead Field Matrix since it returns a matrix of single dipole solutions.

The utility functions are used internally by the functions described above.

The external user configuration can be returned using the NFT_get_config() function. This m-file can be edited manually to specify run-time options for the toolbox. User interface functions call this function to get configuration variables as needed.

The utilbem_compute_cond() and utilbem_compute_indices() functions compute conductivity and index information from the mesh. These functions are called by bem_create_model() and the results are stored in the model structure.

There are two utility functions for computing source (right-hand-side) vectors. utilbem_multilayer_rhs() is used for IPA and utilbem_pot_unbound() is used without IPA.