Real-time tool development for neural dynamic adaptive control
Author(s)
Kate, Sai Pradeep
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Brain activity is inherently noisy and dynamic with hidden brain states affecting our behavior response to external sensory input. With the rapid acceleration of tools developed for monitoring the behavior response and recording neuronal population, we can develop a variety of real-time tools for monitoring animal behavior and controlling neurons. In this study, we have used open source tools to develop two pipelines that can be utilized for observing the behavior response of animals and for closed-loop feedback control of neural populations in real-time. Deeplabcut was used for monitoring the behavior, and the State Space Linear Dynamical Systems approach was used for the dynamic feedback control of neural population. The objective of building both these set-ups is to have low-cost, open source tools that can be used for the adaptive control of neurons under different experimental conditions.
Sponsor
Date
2024-07-27
Extent
Resource Type
Text
Resource Subtype
Thesis