A framework for identification of brain network dynamics using a novel binary noise modulated electrical stimulation...

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Biomedical Engineering

Background:

• Understanding brain network dynamics is crucial for neuroscience and brain-computer interfaces.
• Accurate modeling of brain responses to electrical stimulation is essential for closed-loop control.
• Previous methods for identifying brain network models were limited by input stimulation design.

Purpose of the Study:

• To develop a multi-input-multi-output (MIMO) linear state-space model (LSSM) for brain network dynamics.
• To propose a novel input stimulation design for accurate LSSM identification.
• To improve the precision of mapping electrical stimulation parameters to brain signals like ECoG and EEG.

Main Methods:

• Developed a MIMO linear state-space model (LSSM).
• Designed a novel input signal: a pulse train modulated by binary noise (BN).
• Validated the input design's spectral properties and waveform flexibility through numerical experiments.

Main Results:

• The proposed BN-modulated input pattern meets spectral conditions for accurate system identification.
• This input pattern accommodates desired pulse shapes for electrical stimulation.
• Numerical results demonstrated over 300-fold reduction in relative estimation error compared to other input patterns.

Conclusions:

• The novel binary noise modulated input design significantly enhances the accuracy of LSSM identification for brain network dynamics.
• This method offers a more precise way to model brain responses to electrical stimulation.
• The findings have implications for advancing brain function understanding and closed-loop brain state control.