Related Experiment Video
Updated: Jul 25, 2025
Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models
Published on: August 12, 2018
Combining biophysical models and machine learning to optimize implant geometry and stimulation protocol for
Simone Romeni1, Elena Losanno2, Elisabeth Koert1,3
1Bertarelli Foundation Chair in Translational Neural Engineering, Center for Neuroprosthetics and Institute of Bioengineering, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland.
This study presents an in-silico optimization workflow for peripheral nerve interfaces, improving neuroprosthetic design for motor function restoration. The method enhances selectivity by optimizing implant geometry and stimulation protocols using subject-specific anatomy.
Area of Science:
- Biomedical Engineering
- Neuroscience
- Computational Biology
Background:
- Peripheral nerve interfaces offer potential for restoring lost functions.
- Intraneural interfaces provide high selectivity but depend on precise implantation and anatomy.
- Limited data exists on subject-specific anatomy for optimal interface design.
Purpose of the Study:
- To develop an optimization workflow for pre-surgical planning and stimulation protocols for intraneural electrodes.
- To improve the selectivity and performance of neuroprosthetic devices.
- To reduce computational costs in designing personalized nerve interfaces.
Main Methods:
- Utilized hybrid models (HMs) of neuromodulation and machine learning-based surrogate models.
- Employed particle swarm optimization for in-silico optimization of implant geometry, insertion, and stimulation.
- Incorporated morphological data from the human median nerve.
Main Results:
- Optimized the geometry and number of electrodes for multi-electrode implants.
- Determined optimal electrode insertion and multipolar stimulation protocols.
- Achieved selective activation of all muscles innervated by the human median nerve in silico.
Conclusions:
- Hybrid models are effective for optimizing personalized neuroprostheses for motor function.
- Multipolar stimulation significantly enhances selectivity in peripheral nerve interfaces.
- Subject-specific structural and functional anatomy is crucial for achieving high selectivity in neuroprosthetics.
