Skip to content

Python package for digital phenotyping data synchronization and acquisition

License

Notifications You must be signed in to change notification settings

neurobooth/neurobooth-os

Neurobooth-os

Neurobooth-os is a python package to initialize, synchronize and record behavioral and physiological data streams from wearables, D-/RGB cameras, eye tracker, ECG, mouse and microphone in a booth.

Installations

We recommend the Anaconda Python distribution. To install neurobooth-os, simply do:

$ pip install -e git+https://github.com/neurobooth/neurobooth-os.git#egg=neurobooth_os

and it will install neurobooth-os along with the dependencies which are not already installed.

To check if everything worked fine, you can do:

$ python -c 'import neurobooth_os'

and it should not give any error messages.

Install Pylink running on the STM:

$ pip install --index-url=https://pypi.sr-support.com sr-research-pylink

If pip install from sr-research doesn't work, please follow the following steps:

  • Create an sr-research support account
  • Download the EyeLink Developers Kit v2.1.1 (32 and 64 bit) installer from the SR Research

Support forum. * Install the Eyelink Developers Kit * Run: $ cd C:'\Program Files (x86)\SR Research\EyeLink\SampleExperiments\Python\' * Run: $ python install_pylink.py

For pyspin FLIR installation on the ACQ: Download SDK from https://www.flir.com/products/spinnaker-sdk/?vertical=machine+vision&segment=iis direct link to wheel file: https://flir.app.boxcn.net/v/SpinnakerSDK/file/982785545712 then unzip the file, cd to the folder and run: $ pip install spinnaker_python-2.x.x.x-cp3x-cp3x-win_amd64.whl

Setup

As described below, Neurobooth is designed to run on multiple Windows server machines. To do this, it requires that you setup your servers to communicate via WMI. Please see the documentation (https://github.com/neurobooth/neurobooth-os/blob/master/docs/enable_WMI_instuctions.txt), which explains how to configure WMI for more information.

Neurobooth requires a postgreSQL database running on a server. Connection is established with the function neurobooth_os.iout.metadator.get_conn(). Currently, as specified in ~/.neurobooth_os_secrets the local IP is 192.168.100.1, and remotely it connects to neurodoor.nmr.mgh.harvard.edu using the private key in '~/.ssh/id_rsa'.

To setup a private key, first activate the VPN (partner's virtual private network), then run in the terminal:

$ ssh-keygen
$ ssh-copy-id userID@neurodoor.nmr.mgh.harvard.edu

Next, set up the configuration data. Please see the instructions at: https://github.com/neurobooth/neurobooth-os/blob/master/docs/system_configuration.md

Run

As mentioned above, Neurobooth runs on 3 different computers and the starting point is gui.py. The computers are:

  • CTR (control) computer: This computer hosts the GUI and relays commands

to the other computers to start recording from the Neurobooth devices and presenting stimuli. The lab recorder software is on this computer.

  • STM (stimulus) computer: This computer runs the tasks using psychopy.
  • ACQ (acquisition) computer: This computer acquires data.

Each computer has a server that listens for communication from the other computers in the form of string messages. CTR and STM computer communicate with the database.