Skip to main content

Introduction to ROOKConnect

Summary

ROOKConnect is the core infrastructure designed to authorize, extract, process, and deliver health data from multiple sources via API and SDK. This documentation details its modular architecture, the three-pillar data model, and the delivery mechanisms for secure technical implementations across sandbox and production environments.

ROOKConnect is the central health data integration platform within the ROOK ecosystem. Its primary purpose is to simplify the connection with multiple external providers, including API-based platforms and SDK-based mobile applications, to deliver structured data through a unified model.

The problem it resolves​

Individually integrating multiple health data providers requires managing different API architectures, divergent authentication methods, and heterogeneous response formats.

ROOKConnect acts as an intermediate layer that standardizes the authorization, extraction, normalization, and delivery processes. The result is a consistent and scalable architecture that allows teams to focus on core product development, eliminating the burden of maintaining multiple data pipelines.

Architectural benefits​

Implementing ROOKConnect provides significant technical advantages for the client ecosystem:

  • Broad compatibility: Supports standardized connections to API-based platforms (such as Dexcom, Fitbit, Garmin, Oura, Polar, Whoop, Withings and others) and mobile health kits (such as Apple Health, Health Connect and Samsung Health).
  • Streamlined data management: Transforms raw data into harmonized, standardized, and normalized information, ready to be processed by internal client systems.
  • Flexible delivery options: Prioritizes real-time asynchronous delivery via webhooks, keeping the REST API available for on-demand queries.

General integration flow​

The platform operates through four sequential phases that structure the data lifecycle:

  1. Authorization: The system obtains explicit user consent to access their health data. This process is securely managed through individual endpoints per data provider.
  2. Extraction: Once authorization is confirmed, the platform establishes a secure connection with the corresponding source and retrieves historical and recent data.
  3. Processing: ROOK processes the collected data, applying harmonization and normalization rules under a unified schema to eliminate structural discrepancies between providers.
  4. Delivery: The processed data is transmitted to the client backend, primarily via webhooks.

Data model organization​

ROOKConnect classifies the extracted information into three fundamental pillars:

  • Physical Health: Metrics regarding daily activities, movement, and exercise.
  • Body Health: Physiological variables, body composition, and nutrition.
  • Sleep Health: Metrics on recovery and sleep quality.

Within these three pillars, information is structured into two response formats:

  • Summaries: It represents the collection of data around the day.
  • Events: Granular data points representing specific activities or punctual measurements with exact timestamps.
Data integrity validation

To process both summaries and events correctly, the engineering team must always validate the document_version and datetime fields included in the payload. This ensures the client backend updates records properly and prevents data duplication.

Integration and validation tools​

To accelerate time-to-production, we provide specific tools for different development stages:

Connections Page​

The Connections Page simplifies user authorization during sandbox testing or production. For customized branding and capability, the /authorizer endpoint allows clients to build their own interfaces.

ROOK Extraction APP​

The ROOK Extraction App is a pre-built mobile solution for SDK-based sources such as Apple Health and Health Connect. It facilitates quick deployment without requiring mobile development resources.

Production transition

The Connections Page is designed exclusively for the sandbox environment. For production integrations, the frontend or mobile team must implement the individual authorization endpoint to guarantee total control over the end-user experience and privacy.

Next steps​

With the conceptual architecture clear, the engineering team can proceed to configure the development environments and initiate data ingestion.