Optimization of Stimulation Parameters for Targeted Activation of Multiple Neurons Using Closed-Loop Search Methods

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Neuroscience

Background:

• Differential activation of neuronal populations is key to improving clinical devices like sensory or cortical prostheses.
• Enhancing stimulus specificity is crucial for targeted neuronal activation and biologically realistic percepts.
• Activating neuronal subpopulations with a single electrode requires advanced stimulus optimization techniques.

Purpose of the Study:

• To develop and implement an optimization-based search routine for identifying selective stimulus waveforms.
• To systematically search the stimulus waveform space for targeted neuronal activation.
• To characterize stimulus-evoked neuronal populations and their associated selective stimulus waveform spaces.

Main Methods:

• Applied Powell's conjugate direction method, an optimization routine, to search stimulus waveform space.
• Utilized a 1-D sigmoid activation model and a 2-D strength-duration curve to measure neuronal activation.
• Implemented the search routine in both experimental and simulation studies.

Main Results:

• Demonstrated the ability to activate distinct sub-populations within a larger neuronal group (e.g., 7 sub-populations from 5 neurons).
• Characterized unique stimulus waveform spaces and neuronal activation curves for different neuronal culture and electrode array combinations.
• Identified specific "selectivity spaces" for targeted neuronal activation.

Conclusions:

• The presented optimization method efficiently uncovers neuronal selectivity spaces.
• This approach allows for focused experimental investigation of desired neuronal activation patterns.
• The findings support the development of more precise and effective neural prosthetics.