TIPS（temporal interference parameter simulation）

Authors: Zhang Wei([email protected]) and Ma Ru ([email protected])

This toolbox aims at temporal interference simulation (TIS) optimization with GPU acceleration.
It helps experimenter to evaluate the individual TIS stimulation parameters, including electrode position and current intensity. The evaluation needs the T1 and T2 MRI structural image of the subject to build tetrahedron brain mesh.This step is rely on SIMNIBS pipeline. https://simnibs.github.io/simnibs/build/html/documentation/command_line/headreco.html?highlight=headreco TIPS is based on MATLAB and CUDA.

---

Computer recommended configuration Memory > 8G SSD >= 256G (for virtual memory) Platform: windows 10 GPU: NVIDIA, support CUDA, memory > 4G

---

Installation steps (on Windows 10):

1. Install MSVC 2017 or 2019, then run the following command in MATLAB.

mex -setup

2. Install latest cuda runtime, then run the following command in MATLAB.

gpuDevice

3. Install SIMNIBS and include its MATLAB functions path.
4. Download this toolbox.
5. Run first_AddPath.m to add path to MATLAB.
6. Run first_CompileCUDA.m to compile the cuda script into mexw64 file.
7. For each subject, you could follow headreco to build individual brain mesh. (Or you could run "SIMNIBS_headreco.m".)
8. For each subject, you could follow leadfield to get gray matter middle surface. (Or you could run "SIMNIBS_LF_tri.m".)
9. For each subject, you could follow leadfield to get tetrahdron leadfield. (Or you could run "SIMNIBS_LF_tet.m".)
10. Run TIPSconfig_GUI.mlapp.m to set the optimal parameters. (Or you could use TIPSconfig.m). The configurations are stored in the "cfg" variable.
11. Run OpenTIPS(cfg) in command line Or main.m file to optimal TIs parameters.
12. Run post_plot.m file to plot results.

---

Note:

1. Some plot demo functions are in the folder "ex_Figures".
2. The first time runing optimization for each subject needs more time to prepare input leadfield for GPU. The prepared single float format leadfield is stored under subject folder.
3. The dataRoot variable is the root directory of all the subjects. subMark variable is name or mark of a subject under root directory. The leadfield and orientation (optional) files are stored in the 'TI_sim_result' folder under each subject directory. Each TIS optimal configuration and result is stored under the 'TI_sim_result' folder, with folder name simMark variable.

---

Citation

1. SIMNIBS Saturnino, G. B., Siebner, H. R., Thielscher, A., & Madsen, K. H. (2019). Accessibility of cortical regions to focal TES: Dependence on spatial position, safety, and practical constraints. NeuroImage, 203, 116183.
2. ACID Ruthotto L, Kugel H, Olesch J, Fischer B, Modersitzki J, Burger M, and Wolters C H. Diffeomorphic Susceptibility Artefact Correction of Diffusion-Weighted Magnetic Resonance Images.Physics in Medicine and Biology, 57(18), 5715-5731; 2012, doi: 10.1088/0031-9155/57/18/5715.
3. AAL3 Automated anatomical labelling atlas 3. Rolls, E. T., Huang, C. C., Lin, C. P., Feng, J., & Joliot, M., Neuroimage, 2020, 206, 116189, doi:10.1016/j.neuroimage.2019.116189

---

Lisence

TIPS is copyrighted © by its authors and licensed under GPL v3.