Improving computational models of deep brain stimulation through experimental calibration.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Modeling

Background:

• Deep brain stimulation (DBS) is a key treatment for movement disorders, but understanding stimulation-induced processes is limited.
• Computational models aid insight into DBS but suffer from uncertainties due to lack of validation and calibration.
• Existing models often rely on manufacturer specifications without independent verification.

Purpose of the Study:

• To develop and present a workflow for validating and calibrating computational models of DBS electrodes.
• To enhance the predictive power and reliability of computational models used in DBS research and application.
• To reduce uncertainties in model parameters for more accurate neural activation predictions.

Main Methods:

• A workflow was developed using rodent microelectrodes, incorporating microscopy and in vitro impedance spectroscopy for geometry validation.
• Uncertainties in tissue distribution and dielectric properties were addressed.
• A concept for in vivo impedance spectroscopy was outlined for computational model calibration.

Main Results:

• Electrode characterization revealed significant variability (32.93% std dev in tissue activated volume), highlighting the need for validation.
• The presented workflow demonstrably enhanced the credibility of neural activation predictions in a rodent model.
• The approach provides an accessible method for obtaining validated and calibrated electrode geometries.

Conclusions:

• Reducing model uncertainties significantly increases the accuracy of predicting neural activation.
• The open-source workflow is adaptable for various applications, including human DBS, and allows for model refinement.
• This framework facilitates integration of further experiments for live updates and improved computational models.