Headline

The Aura device will alert an user from an epilepsy seizure within few minutes notice. Aura data analysis platform provides a set of tools to easily acces and analyse anonymized patient data.

Quick Start

• install python 3, pip and virtualenv

• clone the repository

        git clone https://github.com/clecoued/Aura_data_analysis_platform.git

• create a new python virtualenv in the repository (the -p is not necessary if python 3 is your default version)

        cd Aura_data_analysis_platform
        virtualenv myProject -p <pathToPython3>

• install python package dependencies

        pip install -r requirements.txt

• get your personnal credentials file credentials.cfg and copy it to Aura_data_analysis_platform/query/credentials.cfg
• launch the demo script

        cd query
        python queryExample.py

• Success !

API

Database

Data are stored in a two differents databases on Cloud in order to split user personnal information from physiological data. User physiological data remain completely anonymous and data scientists can only access to user UUID information.

User data

User data are stored on a Amazon DynamoDB(NoSql) database.

We acces it using python Amazon library boto3

The database is described as below:

A single table - Users - storing the users list under format:

// User sample
{
    "UUID": "0399a758-da41-4fa5-aa30-625bc19c92ac" //(String) User UUID 
}

Physiological signal data

Physiological signal data are stored on a InfluxDB (Times Series Database) database.

We acces it through:

• REST API - see InfluxDB documentation
• Python InfluxDB client library wrapping the requests

We store informations in a single table - ** physio_signal ** -

We currently save three differents measurements:

Measurement	Measurement name in DB	Recorded data sample	Recorded data description
Heart Rate	heart	rr_interval	(Short)RR-interval in milliseconds
Skin Temperature	temperature	skin_temperature	(Float) Skin temperature in degree Celsius with a precision 0.01
Electro Dermal Activity	electro_dermal_activity	output_sensor_frequency	(Integer) Output sensor frequency in Hz
		electro_dermal_activity	(Integer) Electro dermal activity in Ohm

Each measurement is tagged with following fields and queries can be filtered along theses tags:

TAG	Description
uuid	(String) sample UUID
user	(String) user UUID see User data section
type	(String) data type / RrInterval / SkinTemperature / ElectroDermalActivity /
device_adress	(String) device adress UUID
time	date and timestamp in milleseconds

Data Viz