neurone-tools 1.1

Binary file reader for NeurOne Tesla EEG data

pyneurone_tools

A simple python library to read Bittium NeurOne Tesla EEG system binary data

Function overview

Basic Usage

from pyneurone_tools import neurone_tools
from pathlib import Path

session_path = Path('./exciting_project/exciting_data/2022-01-03T090018')
recording_id = 1

nt = neurone_tools(session_path, recording_id, channels='all')

loading data

data, channel_names = nt.load_data()

The data has the shape (samples, channels), unit is nanovolt.

loading events

events = nt.load_events()

Evens are returned as a pandas DataFrame.
The columns are:

revision: Event marker revision number
type: Type of event marker
source_port: Source port number
channel_number: Channel used for event marking
8bit_trigger_code: The 8-bit TTL trigger code if the marker has one, otherwise None
start_sample_index: sample point where trigger occured
stop_sample_index: sample point where trigger ended
description_length: length of event comment (string in recording system)
description_offset: offset of event comment (string in recording system)
data_length: length of additional data of event (any binary data)
data_offset: offset of additional data of event (any binary data)
source_port_name: name of the channel from channel_number
description: name of the event marker type

sampling rate

fsample = nt.get_sampling_rate()

loading channel names

channel_names = nt.channels_avail

loading alternating current (AC) mode flags

ac_mode = nt.get_ac_mode()

loading filter parameters

filter_settings = nt.get_filter_settings()

Terminology

The following terms are used in this package

Session

A session is the top-level term, think of it as the folder containing all the other files.
It is started when you click 'Start' in the recording software and can be made up from several recordings.

Recording

A recording is what you get when you press the 'Record' button and ends when you click the 'Stop Recording' button.
Recording ID numbers are 1-indexed, as defined by the recording software.

Binary File

A Recording can be split into several Binary files, as one can be specified in the NeurOne recording software. Binary file IDs are 1-indexed, as this is given by the recording software.
It is unikely that the user would ever need to access these files directly.

Installation

Supported Python Versions

This code is tested on python versions 3.7, 3.8, 3.9, and 3.10.
Earlier versions could be usable, but need an older version of pandas.

Using pip

pip install neurone_tools

pip directly from github

pip install git+https://github.com/jkschluesener/pyneurone_tools.git@master

pip without PyPi

cd <your_preferred_code_folder>
git clone https://github.com/jkschluesener/pyneurone_tools.git
pip install ./pyneurone_tools/

Development mode

Link this repo's folder into your current python with the -e (external) flag.

cd <your_preferred_code_folder>
git clone https://github.com/jkschluesener/pyneurone_tools.git
pip install -e ./pyneurone_tools/

or use ssh if you prefer

cd <your_preferred_code_folder>
git clone [e-mail]:jkschluesener/pyneurone_tools.git
pip install -e pyneurone_tools

Changes to this directory will be available without reinstall.
Upon local changes of the code, you may have to re-import this package or maybe restart your python kernel.

Adding functionality

Is there a funcitonality you would like to see in this package?
Just open an Issue for us / me to discuss and implement it.

License

This code is GPLv3 licensed.

Layout of the binary files was taken from the official NeurOne manual.
This code is also in parts derived from the original Bittium / Mega Electronics Ltd matlab toolbox (official / mirror) to map integer ID codes to their string counterparts.