Automated Orchestration of Observational Research

Our client was a U.S.-based, multi-institutional healthcare research initiative developing an AI-ready clinical data ecosystem for critical-care research. The initiative brought together several institutional data sources with differences in clinical workflows, source systems, coding practices, data availability, and OMOP ETL implementations. The client needed a systematic way to translate complex clinical research questions into reproducible computable phenotypes that could support cohort discovery, population characterization, outcome definition, and downstream analytical and machine-learning workflows. The project focused on structured electronic health record data represented primarily through the OMOP Condition, Procedure, Measurement, Observation, Device, Drug, Specimen and Visit domains. To preserve confidentiality, the organization and individual clinical use cases are not disclosed.

Our client, a U.S. health insurer collaborating with multiple hospital systems, aimed to aggregate and harmonize anonymized claims and clinical data in the PCORnet Common Data Model (CDM) to support large-scale outcomes research and operational analytics. The incoming medical and billing feeds came from heterogeneous hospital and payer systems with inconsistent schemas, variable data quality, and no unified governance. The client asked SciForce to design and implement a sustainable, cloud-native ETL/ELT pipeline on Snowflake that would: 1) Continuously integrate raw source feeds into a centralized Snowflake data platform; 2) Transform them into a PCORnet-conformant CDM with strong data quality guarantees; 3) Enable near real-time analytics for patient demand forecasting, capacity planning, and revenue cycle optimization.

The client was a residential care facility for elderly patients who were at risk of falling, relying on wearable sensor fall detection for assisted living facilities to reduce incidents and response time. Each resident wore a small motion-tracking device as part of a wearable health monitoring setup that measured their movements in three directions. Their rooms were also equipped with AI in assisted living facilities that monitored temperature, humidity, light, noise levels, and whether someone was in or out of bed. The goal was to build an AI fall detection system that could automatically detect when someone had fallen and notify staff quickly. The team focused on understanding how movement patterns could indicate a fall and explored how room data could help add context in the future.

Our client is a university hospital based in Germany, aimed to enhance cross-institutional standardized healthcare data exchange through structured health data pipelines. Multiple institutions involved in observational research sought integrating observational research data into clinical systems via FHIR, including risk models and disease prevalence metrics, into operational clinical workflows. To support this goal, the client required a conversion pipeline from OMOP CDM (used in research analytics) to HL7 FHIR (used in clinical applications) to enable real-time data AI for healthcare data interoperability.