Overview

The Thryve Health Analytics Platform processes wearable sensor data into health metrics following clinical and scientific best practices in near real-time. Our analytics engine transforms data from hundreds of data signals into standardized health insights suitable for healthcare applications, research, and population health monitoring.

All analytics are computed and available within milliseconds of data ingestion, enabling immediate access to health insights. Algorithms are based on established clinical frameworks and peer-reviewed research, ensuring that metrics meet healthcare application requirements.

Core Analytics Capabilities

Real-Time Processing

Analytics are calculated continuously as new data arrives, rather than through batch processing. This enables immediate health insights and supports applications requiring real-time health monitoring.

Scientific backing

The analytics platform is built on our roots in biosignal analysis and our scientific approach given Thryve was built as a spin-off of renowned Fraunhofer Institute. Where applicable we build our algorithms on top of established clinical frameworks adapted for continuous wearable data monitoring and are actively involved in research projects all around Europe.

Data Harmonization

Our data harmonization efforts go beyond storing and providing data from different kinds of data sources in a unified data format and API. With our extensive analytics suit we harmonize core health metrics across different wearable manufacturers and data sources. This solves the fundamental challenge of inconsistent metric definitions (e.g. for sleep data) between devices and enables longitudinal health tracking regardless of device changes.

Multi-Device/Source Data Fusion

Advanced algorithms handle data from multiple simultaneous sources:

• Overlap Detection: Identifies and resolves conflicts when multiple devices record the same timeframe

• Priority-Based Selection: Hierarchical data selection when conflicts arise

• Data Deduplication: Prevents double-counting across connected devices

Data priority is determined based on information on recording device and context.