Human DBS Example

This example demonstrates DBS modeling on a human subject using MRI- and DTI-based tissue characterization and a Boston Scientific Vercise directed electrode. It illustrates how to configure anatomical data, electrode orientation and placement, dielectric modeling, solver settings, and typical output options.

This example is representative of patient-specific DBS simulations based on real neuroimaging data.

Input Highlights

• BrainRegion: spherical subvolume around the stimulation site, sufficiently large to include local tissue heterogeneity and reduce boundary effects.

• Electrode: Boston Scientific Vercise directed lead with its true shaft orientation in the patient’s anatomical coordinate system.

• MRI: used to distinguish CSF, GM, WM, and blood.

• DTI: provides local anisotropy (tensor conductivity).

• Dielectric model: 4-Cole-Cole dispersion model (standard for brain tissue).

• Mesh: adaptive strategy with fine resolution near the stimulation region.

• Stimulation: frequency-domain solution at 10 kHz using a multisine formulation.

• Outputs: impedance estimate, electrode geometry, and field/VTP data exported for visualization and post-analysis.

Full Input Example

{
  "BrainRegion": {
    "Center": {"x[mm]": -13.99, "y[mm]": -7.73, "z[mm]": -7.91},
    "Dimension": {"x[mm]": 70.0, "y[mm]": 70.0, "z[mm]": 70.0},
    "Shape": "Sphere"
  },
  "Electrodes": [
    {
      "Name": "BostonScientificVerciseDirected",
      "CustomParameters": null,
      "Rotation[Degrees]": 0.0,
      "Direction": {"x[mm]": -0.45, "y[mm]": 0.65, "z[mm]": 0.61},
      "TipPosition": {"x[mm]": -13.99, "y[mm]": -7.73, "z[mm]": -7.91}
    }
  ],
  "MaterialDistribution": {
    "MRIPath": "./input_files/sub-John_Doe/JD_segmask.nii.gz",
    "MRIMapping": {
      "Unknown": 0,
      "CSF": 3,
      "White matter": 2,
      "Gray matter": 1,
      "Blood": 4
    },
    "DiffusionTensorActive": true,
    "DTIPath": "./input_files/sub-John_Doe/JD_DTI_NormMapping.nii.gz"
  },
  "DielectricModel": {"Type": "ColeCole4", "CustomParameters": null},
  "Mesh": {
    "LoadMesh": false,
    "LoadPath": "",
    "MeshingHypothesis": {
      "Type": "Fine",
      "MaxMeshSize": 1000.0,
      "MeshSizeFilename": ""
    },
    "SaveMesh": false
  },
  "StimulationSignal": {
    "CurrentControlled": false,
    "Type": "Multisine",
    "ListOfFrequencies": [10000.0]
  },
  "Solver": {
    "Type": "CG",
    "Preconditioner": "bddc",
    "PreconditionerKwargs": {},
    "PrintRates": true,
    "MaximumSteps": 200,
    "Precision": 1e-8
  },
  "PointModel": {
    "Pathway": {
      "Active": false,
      "FileName": ""
    },
    "Lattice": {
      "Active": true,
      "Center": {"x[mm]": -13.99, "y[mm]": -7.73, "z[mm]": -7.91},
      "Shape": {"x": 20, "y": 20, "z": 20},
      "Direction": {"x[mm]": -0.45, "y[mm]": 0.65, "z[mm]": 0.61},
      "PointDistance[mm]": 0.5
    }
  },
  "OutputPath": "Results",
  "ComputeImpedance": true,
  "ExportVTK": true,
  "ExportElectrode": true
}

Explanation of Key Parameters

BrainRegion: A 70 × 70 × 70 mm spherical region centered at the targeted stimulation site. This ensures: - inclusion of relevant neuroanatomy, - realistic current spread, - minimized artificial boundary effects.

Electrode Orientation: The Direction vector aligns the shaft with the patient’s anatomy. This supports: - directional stimulation, - contact-wise modeling, - and consistent mapping with tractography.

Tissue Properties (MRI + DTI): - MRI segmentation defines local tissue class assignment (GM, WM, CSF…). - DTI maps anisotropic conductivity tensors into OSS-DBSv2.

Dielectric Model: Cole-Cole dispersion (4-term) is the standard frequency-dependent model for brain tissue. No custom override is used here to preserve realistic electrical behavior.

Meshing: Fine hypothesis ensures adequate resolution in the vicinity of the electrode. The mesh is generated automatically.

Stimulation: - Frequency-domain solution at 10 kHz. - Multisine allows solving a single frequency slice efficiently.

Outputs: - Impedance provides model diagnostics and electrode–tissue interface estimation. - VTK files allow visualization (ParaView, Lead-DBS, MNE, etc.). - The electrode geometry is exported for debugging and high-quality plotting.

Running the Example

Execute the simulation from the command line:

ossdbs input.json

A logfile and the result files (VTK, impedance, and optional outputs) will be written to the Results/ folder.

Typical Uses

This type of patient-specific setup is used for:

• evaluating contact selection,

• comparing different stimulation parameters,

• impedance analysis,

• grounding and reference validation,

• tract-related field interpretation,

• pre/post clinical electrode position review.