Computational Models of Ultrasound Neuromodulation

Performance site: Campus Biotech, Geneva

Background

Ultrasound stimulation (US) has recently emerged as a promising technology to achieve reliable, selective and noninvasive neuromodulation of various targets of the central nervous system of rodents, non-human primates and humans (King et al., 2013; Legon et al., 2014). A myriad of applications can therefore be envisaged in which US would replace the standard and invasive electrical stimulation. However, depiste the fast emergence of this scientific field, to date, the fundamental mechanism(s) by which ultrasonic waves can modulate neural activity are still unkown. This is a limiting factor hindering the maturation of US as a reliable neuromodulation technology.

We are currently developing computational modeling framework to decipher the mechanisms of Ultrasound Neuromodulation. This framework is based on the recently proposed biophysical mechanism of intramembrane cavitation (Krasovitski et al., 2011; Plaksin et al., 2014). The projects presented below include the development of various branches of this computational framework, as well as its validation against experimental data.

Projects

Project description: Part of our computational framework development includes the implementation of our models in the NEURON model specification language (NMODL), and their integration within our core framework developed in Python. We have started to implement “point-neuron” models (i.e. no spatial extent) enabling this Python-NEURON communication. The student’s main task will be to extend these point-neuron models to multi-compartmental, morphologically realistic neuron representations.

Activities:

  • Familiarize with the point-neuron model, underlying differential equations, and the current Python+NEURON modeling pipeline
  • Develop innovative strategies to build and connect multiple compartments in NEURON within the very singular frame of our “intramembrane cavitation” based model
  • Implement and validate a generic model consisting of one soma connected to multiple axons and dendrites
  • Adapt the generic model to specific neuron types

Requirements:

  • Basic knowledge of Hodgkin-Huxley models / differential equations
  • Basic programming skills
  • Experience with Git, Python, NEURON and C is a plus

Best for: semester / master project (to be discussed)

Contact: [email protected]

References:

  • Lemaire, T., Neufeld, E., Kuster, N., and Micera, S. (2019). Understanding ultrasound neuromodulation using a computationally efficient and interpretable model of intramembrane cavitation. J. Neural Eng.
  • Plaksin, M., Shoham, S., and Kimmel, E. (2014). Intramembrane Cavitation as a Predictive Bio-Piezoelectric Mechanism for Ultrasonic Brain Stimulation. Phys. Rev. X 4.
  • Hines, M., Davison, A.P., Muller, E., Hines, M.L., Davison, A.P., and Muller, E. (2009). NEURON and Python. Front. Neuroinform. 3, 1.

Project description: Part of our computational framework development includes the validation of our models against experimental data. So far, we have implemented “point-neuron” models of several generic cortical neuron types, in Python. However, most of our experimental data concerns the responses of sensory neurons of the medicinal leech, for which we don’t have models yet. The student’s main task will be to adapt our point-neuron models to these specific leech neurons, and then to conduct a thorough quantitative comparison between model predictions and experimental data (already) acquired on those neurons.

Activities:

  • Familiarize with the point-neuron model, underlying differential equations, and the current Python implementation
  • Search the literature to identify a clear nomenclature of key ion channel populations present in the targeted cell types (touch, pressure and nociceptive cells), along with the underlying gating dynamics of these ion channels (Hodgkin-Huxley equations)
  • Implement those point-neuron models and validate them against observations from the literature (spontaneous activity, response to electrical stimulation, …)
  • Establish key features of neural responses to ultrasound stimulation predicted by these models
  • Conduct a thorough comparison of model predictions with experimental data, using relevant metrics
  • Write a report with presentation and discussion of results

Requirements:

  • Basic knowledge of Hodgkin-Huxley models / differential equations
  • Basic programming skills
  • Experience with Git and Python is a plus

Best for: semester / master project (to be discussed)

Contact: [email protected]

References:

  • Lemaire, T., Neufeld, E., Kuster, N., and Micera, S. (2019). Understanding ultrasound neuromodulation using a computationally efficient and interpretable model of intramembrane cavitation. J. Neural Eng.
  • Plaksin, M., Shoham, S., and Kimmel, E. (2014). Intramembrane Cavitation as a Predictive Bio-Piezoelectric Mechanism for Ultrasonic Brain Stimulation. Phys. Rev. X 4.
  • Johansen, J. (1991). Ion conductances in identified leech neurons. Comp Biochem Physiol A Comp Physiol 100, 33–40.

Project description: Part of our computational framework development includes the coupling (1) morphologically detailed, biophysical neuron models and (2) models of acoustic propagation within realistic neural structures (brain, spinal cord, peripheral nerves) and their anatomical environment. The student’s main task will be to develop a model of acoustic propagation within the peripheral nerve and its anatomical environment, from a distant ultrasound transducer, using the Sim4Life simulation platform (https://zmt.swiss/sim4life/).

Activities:

  • Familiarize with the different bioeffects of ultrasound, the theory of acoustic propagation and the main existing mathematical models (Kyriakou, 2015)
  • Search the relevant literature for acoustic propagation properties of different nerve tissues
  • Familiarize with the Sim4Life simulation platform and the main modeling workflow
    • Nerve segmentation from imaging data
    • Definition of the model geometry
    • Assignment of material properties
    • Mesh creation, model discretization
    • Simulation
    • Results analysis
    • Automation using Python scripts
  • Develop and validate a model of acoustic propagation within a peripheral nerve and its environment
  • Define and implement metrics to quantify the distribution of acoustic intensity within the nerve
  • Compare the performances of different transducer types and geometries

Requirements:

  • Basic knowledge of Finite Element Models (FEM) / Finite Element Analysis (FEA)
  • Basic programming skills
  • Experience with Python is a plus

Best for: master project

Contact: [email protected]

References:

  • Kyriakou (2015). Multi-Physics Computational Modeling of Focused Ultrasound Therapies.

Project description: Our computational framework currently includes models of several generic different neuron types, but that list is far from exhaustive. Moreover, the addition of new neuron types to the framework is still cumbersome, as it requires manual implementation. The student’s main task will be to develop tools for automatic import and export of neuron models between our framework and the NeuroML model specification language (Gleeson et al., 2014, https://www.neuroml.org/). This task is a critical towards the acceptance of the framework by the broader scientific community.

Activities:

  • Familiarize with the point-neuron model, underlying differential equations, and the current Python-NEURONmodeling pipeline
  • Familiarize with the NeuroML specification language
  • Develop and validate an import tool to generate a neuron Python class from a NeuroML model specification
  • Develop and validate an export tool to generate a MOD file that can then be used to compile a membrane mechanism in the NEURON simulation environment.
  • Use the import tool to further populate the package with additional neuron types

Requirements:

  • Basic knowledge of Hodgkin-Huxley models / differential equations
  • Basic programming skills
  • Experience with Git, Python, NEURON and C is a plus

Best for: semester project

Contact: [email protected]

References:

  • Lemaire, T., Neufeld, E., Kuster, N., and Micera, S. (2019). Understanding ultrasound neuromodulation using a computationally efficient and interpretable model of intramembrane cavitation. J. Neural Eng.
  • Hines, M., Davison, A.P., Muller, E., Hines, M.L., Davison, A.P., and Muller, E. (2009). NEURON and Python. Front. Neuroinform. 3, 1.
  • Gleeson, P., Crook, S., Cannon, R.C., Hines, M.L., Billings, G.O., Farinella, M., Morse, T.M., Davison, A.P., Ray, S., Bhalla, U.S., et al. (2010). NeuroML: A Language for Describing Data Driven Models of Neurons and Networks with a High Degree of Biological Detail. PLOS Computational Biology 6, e1000815.

Contact

If none of the projects suit you but you are interested in Ultrasound Neuromodulation or computational models in general, please feel free to contact us to discuss potential opportunities.

Théo Lemaire ([email protected])