Wearable Health Data Normalization Pipeline

Design a data normalization microservice that can ingest health data from multiple wearable devices (Fitbit, Apple Watch, Garmin) and standardize metrics into a unified database schema. Create a flexible transformation layer that can handle varying data formats, sampling rates, and sensor types. Implement machine learning-powered data quality assessment to detect and correct sensor anomalies. Develop a real-time streaming architecture that can process thousands of device streams simultaneously.