In order to perform statistical analysis and train machine learning models that are able to find patterns and predict physiological responses from the data, we need to create data repositories. The data is produced by studies (with or without humans) that incorporate various devices for different purposes (e.g., for use cases such as predicting asthma exacerbation, or just to evaluate the quality and reliability of devices at extracting biomarkers or tracking environmental factors). The data produced by these studies needs to be kept in a consistent format and stored in an accessible database for anyone partner to access.
Data produced by the studies is been stored into a database using secured cloud resources. The upload of the information takes place through encrypted channels either at the end of a recording session or by real-time streaming and processing. All data is deidentified to ensure privacy before sharing with our partners and the scientific community.