Zeffiro Forward and Inverse Interface for Complex Geometries

Zeffiro Interface© 2018- Sampsa Pursiainen & ZI Development TeamIntroductionZeffiro Interface (ZI) is an open source code package constituting an accessible tool formultidisciplinary finite element (FE) based forward and inverse simulations in complex geometries.Developed for MATLAB, ZI aims to streamline the process of analyzing brain activity and make itaccessible to researchers and clinicians alike. With ZI, one can generate a volumetric finiteelement mesh for a realistic multilayer geometry, such as human brain or an asteroid.A suitable surface segmentation can be produced, for example, usingthe FreeSurfer software suite (Copyright © FreeSurfer, 2013). ZI allows importing a parcellation created with FreeSurfer to enabledistinguishing different brain regions and, thereby, analysing the sub-cortical structures andconnectivity of the brain function over a time series. In each compartment, the orientation of theactivity can be either normally constrained or unconstrained. The main routines of ZI can beaccelerated significantly in a computer equipped with a graphics computing unit (GPU). It isespecially recommendable to perform the forward simulation process, i.e., to generate the finiteelement mesh, the lead field matrix and to interpolate between different point sets, utilizing aGPU. After the forward simulation phase, the model can be processed also without GPU acceleration.RequirementsMATLAB >= r2019aToolboxesSignal Processing ToolboxOptimization toolboxParallel Computing ToolboxStatistics and Machine Learning ToolboxGPU (optional)InstallationZeffiro Interface relies on the external repositories, such as SDPT3, SeDuMi and others, that willbe fetched automatically when cloning with --recurse-submodules. See the installation wiki pagefor instructions on how to install Zeffiro Interface.Getting StartedTo start ZI, open Matlab, navigate to the project directory and run the following command:zeffiro_interface()To learn more about startup options, run one of the following commands in Matlab:help zeffiro_interface% ordoc zeffiro_interfaceRead the wiki pages for more information.ContributingSee CONTRIBUTING.md for details on how to contribute to the project.LicenseZeffiro Interface is licensed under the GNU GPLv3.See the LICENSE file for details.Submodules from external directory are licensed as follows:CVX: GNU GPLv3fieldtrip: GNU GPLv3OSQP: Apache v2.0SDPT3: GNU GPLv2SeDuMi: GNU GPLv2SESAME: unknownspm12: GNU GPLv2Related papersThe interface itself has been introduced in:He, Q., Rezaei, A. & Pursiainen, S. (2019). Zeffiro User Interface for Electromagnetic BrainImaging: a GPU Accelerated FEM Tool for Forward and Inverse Computations in Matlab.Neuroinformatics, https://doi.org/10.1007/s12021-019-09436-9Recent papers:Galaz Prieto, F., Rezaei, A., Samavaki, M., & Pursiainen, S. (2022). L1-norm vs. L2-norm fittingin optimizing focal multi-channel tES stimulation: linear and semidefinite programming vs.weighted least squares. Computer Methods and Programs in Biomedicine, 226,107084, https://doi.org/10.1016/j.cmpb.2022.107084Lahtinen, J., Koulouri, A., Rezaei, A., & Pursiainen, S. (2022). Conditionally Exponential Priorin Focal Near-and Far-Field EEG Source Localization via Randomized Multiresolution Scanning (RAMUS). Journal of Mathematical Imaging and Vision,1-22. https://doi.org/10.1007/s10851-022-01081-3Rezaei, A., Lahtinen, J., Neugebauer, F., Antonakakis, M., Piastra, M. C., Koulouri, A., Wolters,C. H., & Pursiainen, S. (2021). Reconstructing subcortical and cortical somatosensory activity viathe RAMUS inverse source analysis technique using median nerve SEP data. NeuroImage, 245, 118726.https://doi.org/10.1016/j.neuroimage.2021.118726Rezaei, A., Koulouri, A., & Pursiainen, S. (2020). Randomized multiresolution scanning in focaland fast E/MEG sensing of brain activity with a variable depth. Brain Topography, 33(2),161-175. https://doi.org/10.1007/s10548-020-00755-8The essential mathematical techniques used in the interface have been reviewed and validated in:Miinalainen, T., Rezaei, A., Us, D., Nüßing, A., Engwer, C., Wolters, C. H., & Pursiainen, S. (2019). A realistic, accurate and fast source modeling approach for the EEG forward problem.NeuroImage, 184, 56-67. https://doi.org/10.1016/j.neuroimage.2018.08.054Pursiainen, S. (2012). Raviart–Thomas-type sources adapted to applied EEG and MEG: implementationand results. Inverse Problems, 28(6), 065013. https://doi.org/10.1088/0266-5611/28/6/065013The IAS MAP (iterative alternating sequential maximum a posteriori) inversion method and thehierarchical Bayesian sampler are based on:Calvetti, D., Hakula, H., Pursiainen, S., & Somersalo, E. (2009). Conditionally Gaussianhypermodels for cerebral source localization. SIAM Journal on Imaging Sciences, 2(3),879-909. https://doi.org/10.1137/080723995It has been applied for a realistic brain geometry, e.g., in:Lucka, F., Pursiainen, S., Burger, M., & Wolters, C. H. (2012). Hierarchical Bayesian inferencefor the EEG inverse problem using realistic FE head models: depth localization and sourceseparation for focal primary currents. Neuroimage, 61(4),1364-1382. https://doi.org/10.1016/j.neuroimage.2012.04.017The current preserving source model combines linear (face-intersecting) and quadratic (edgewise)elements via the Position Based Optimization (PBO) method and the 10-source stencil in which 4 facesources and 6 edge sources are applied for each tetrahedral element containing a source:Bauer, M., Pursiainen, S., Vorwerk, J., Köstler, H., & Wolters, C. H. (2015). Comparison study forWhitney (Raviart–Thomas)-type source models in finite-element-method-based EEG forward modeling.IEEE Transactions on Biomedical Engineering, 62(11),2648-2656. https://doi.org/10.1109/TBME.2015.2439282Pursiainen, S., Vorwerk, J., & Wolters, C. H. (2016). Electroencephalography (EEG) forwardmodeling via H (div) finite element sources with focal interpolation. Physics in Medicine &Biology, 61(24), 8502. https://doi.org/10.1088/0031-9155/61/24/8502Zeffiro Interface is not intended for use in clinical applications. The authors do not assumeresponsibility for the results obtained with ZI when using clinical data.