The Neural Data Simulator is a real-time system for generating electrophysiology data from behavioral data (e.g. cursor movement, arm kinematics, etc) in real-time. The NDS system can be used to test and validate closed-loop brain-computer interface systems without the need for a human in the loop, generate synthetic data for algorithm optimization, and provide a platform on which to develop BCI decoders.
See the documentation for a complete system overview, installation instructions, and API details.
Ensure that Python >=3.9
and <3.12
is installed. Then, proceed to install LSL:
# on Linux/WSL
conda install -c conda-forge liblsl
# on macOS
brew install labstreaminglayer/tap/lsl
# on Windows
# should be installed automatically by pip when installing NDS
Install the NDS package with the included examples and utilities via pip:
pip install "neural-data-simulator[extras]"
Run the following scripts:
nds_post_install_config
run_closed_loop
NOTE: You might need to give permissions like network access when running the scripts.