#Overview

Cloudbrain is a platform for real-time sensor data analysis and visualization.
One type of sensor data that works well with cloudbrain is EEG data. EEG is the recording of electrical activity along the scalp. In other words, brainwaves.

Cloudbrain enables you to:

• Stream sensor data into a central database.
• Analyze sensor data to find spatial and temporal patterns.
• Visualize sensor data and patterns in real-time.

This makes use of the demo version of cloudbrain running at demo.cloudbrain.rocks.

• Publishers send data to cloudbrain.
• Subscribers receive data from cloudbrain.

On Linux, install the liblo package.

If you're on OSX, make sure you have homebrew and the OSX command line utilities installed Then install liblo through brew:

• brew install liblo

Then install cython and flask through pip:

• pip install cython
• pip install flask

• git clone https://github.com/marionleborgne/cloudbrain.git
• cd cloudbrain

• python setup.py install

You may have to add sudo in front (sudo python setup.py install) if it says "Permission denied".

• Run python cloudbrain/publishers/sensor_publisher.py
• Use the --help flag for the docs.

You can stream some Muse mock data using:

• python cloudbrain/publishers/sensor_publisher.py --mock -n muse -i octopicorn

• Write data to a file: python cloudbrain/subscribers/file_writer_subscriber.py
• Use the --help flag for the docs.

For the mock data streamed above, the command would be:

• python cloudbrain/subscribers/file_writer_subscriber.py -i octopicorn -n muse -m eeg

• python setup.py install

• Install RabbitMQ and start it.
• Create cloudbrain user: rabbitmqctl add_user cloudbrain cloudbrain
• Grant permissions: rabbitmqctl set_permissions cloudbrain ".*" ".*" ".*"

• Run python cloudbrain/publishers/sensor_publisher.py
• Use the --help flag for the docs.

• Write data to a file: python cloudbrain/subscribers/file_writer_subscriber.py
• Use the --help flag for the docs.

• Install Cassandra and start it.
• cd cloudbrain/datastore
• Generate Cassandra schema: python generate_cassandra_schema.py
• Execute schema: bin/cqlsh -f <path_to_cassandra_schema>/cassandra_schema.cql
• Start storing data: python cloudbrain/subscribers/cassandra_pika_subscriber.py

• cd cloudbrain/ui
• npm install
• bower install
• Start the REST API server python cloudbrain/datastore/rest_api_server.py (Hosted version will be used if not started)
• Run gulp to start live-reloading local ui server.
• Use the cloud brain ui in the browser window that opens.

• Install PyInstaller pip install pyinstaller
• For the publishers: cd cloudbrain/publishers/bin/
• Make the scripts executable chmod +x *.sh
• On OSX, generate OSX binaries: sh package_publisher_osx.sh
• On Ubuntu, generate Ubuntu binaries: sh package_publisher_ubuntu.sh
• For the Subscribers cd cloudbrain/subscribers/bin/
• Make the scripts executable chmod +x *.sh
• On OSX, generate OSX binaries: sh package_subscriber_osx.sh
• On Ubuntu, generate Ubuntu binaries: sh package_subscriber_ubuntu.sh

For this example, let's use the demo API server demo.apiserver.cloudbrain.rocks, the device name muse and the device id octopicorn.

• Raw EEG for the OpenBCI: GET http://demo.apiserver.cloudbrain.rocks/data?device_name=openbci&metric=eeg&device_id=octopicorn
• Raw EEG for the Muse: GET http://demo.apiserver.cloudbrain.rocks/data?device_name=muse&metric=eeg&device_id=octopicorn
• Mellow metric for the Muse: GET http://demo.apiserver.cloudbrain.rocks/data?device_name=muse&metric=mellow&device_id=octopicorn
• Concentration metric for the Muse: GET http://demo.apiserver.cloudbrain.rocks/data?device_name=muse&metric=concentration&device_id=octopicorn
• Etc...

Power Bands for the Muse. Get alpha, beta, gamma, theta, delta values all at once, for the same timestamp.

• GET http://demo.apiserver.cloudbrain.rocks/power_bands?device_name=muse&device_id=octopicorn

Get the list of device IDs IDs that are publishing to cloudbrain

• GET http://demo.apiserver.cloudbrain.rocks/registered_devices

Get the list of device names supported by cloudbrain

• GET http://demo.apiserver.cloudbrain.rocks/device_names

CloudBrain was in use at the Exploratorium as part of the Exhibit called Cognitive Technologies.

William wrote an excellent post about our work for the the Exploratorium Exhibit.

All the EEG headsets in the exhibit are sending data to CloudBrain. This data is being routed to booths where visitors can control different things with their brain. For visitors who are willing to share their data, CloudBrain computes aggregates and displays a baseline of the average brain. On the central screen, visitors can see everyone else's live EEG data. Each radar chart shows the state of the main brainwaves (alpha, beta, theta, gamma, delta). This is particularly interesting to see how one's brain compares to others, or to understand how it reacts to different stimuli.

Over 30 people played a role in the project, many of them building insanely awesome booths connected to CloudBrain: 3D-printed lighted flowers, Virtual Reality rock levitation, EEG/Heart Rate Variability correlation, and EEG reactive light tables, a brain-controlled robotic arm, 3D brain reconstructions, and fMRI algorithms. It. Was. Awesome.

Many thanks to my teamates:

• William
• James
• David