IBM Watson Health announces the IBM FHIR® Server to enhance health data interoperability

By | 2 minute read | October 28, 2019

Creative businessman looking at white board

Interoperability in healthcare data is critical to improving clinical and operational workflows. Without it, health organizations are challenged to extract value from data to achieve simpler processes, better care, improved experiences, and faster breakthroughs.

Today, IBM Watson Health is advancing health data interoperability with the announcement of the IBM FHIR® Server to the developer community. The IBM FHIR Server is an open-source Java implementation of the HL7 FHIR® R4 specification that has passed the AEGIS Touchstone “basic” test suite for FHIR R4 with 100 percent conformance.

FHIR (Fast Healthcare Interoperability Resources), which is becoming the standard for interoperability in healthcare, describes the APIs and data formats for exchanging health information. By reducing friction in the flow of healthcare data, the IBM FHIR Server will enable health stakeholders to more easily apply data for more advanced capabilities such as machine learning, to ultimately improve decision making by creating interoperability between applications that were previously incompatible.

IBM Watson Health was an early adopter of the HL7 FHIR specification, beginning own server implementation in 2016 and using this server across the portfolio in a wide range of use cases, including:

  • FHIR-native operational data store: mobile applications use the FHIR HTTP API to store and retrieve FHIR resources in IBM DB2 (e.g. clinical or commercial study participant data)
  • Real-time data ingestion: clinical studies are on-boarded using FHIR implementation-defined operations and the FHIR batch/transaction API is used to ingest large quantities of data from health devices, feeding Kafka and a data lake which is used for downstream analytic applications
  • Longitudinal patient record: clinical data is ingested from multiple sources, normalized, transformed into FHIR, and exposed via the FHIR HTTP API for use in end-user applications (e.g. Value-Based Care)
  • Oncology: A profiled flavor of FHIR resources are ingested, validated, annotated, and exposed for use in Oncology solutions, including Watson for Oncology and Watson Clinical Trial Matching

Pledge to interoperability

The release of the IBM FHIR Server under an Apache 2.0 license is the next step in IBM’s commitment to health data interoperability. IBM supports the efforts of the Office of the National Coordinator for Health Information Technology and Centers for Medicare & Medicaid Services to advance safe and secure consumer-centric information sharing across the health ecosystem through open APIs, while continuing to advocate for trust, transparency, and openness.

Last year, IBM joined leaders from across the industry at the CMS Blue Button 2.0 Developer Conference to pledge its commitment to health data interoperability and frictionless exchange. This summer, IBM renewed this pledge and enhanced FHIR Server implementation with support for version R4, the first release of the specification with normative content.

Releasing this work under a permissive open source license is the next step on IBM’s journey to help clients drive digital transformation. Last month, Forrester named IBM a Leader in “The Forrester Wave™: Enterprise Health Clouds, Q3 2019,” specifically calling out:

“IBM has strategically designed a cloud platform that differentiates with its ability to solve healthcare-specific problems, such as native FHIR and DICOM integration, patient matching, and risk stratification.”

These proven capabilities around cloud, data, and analytics, combined with deep industry expertise and commitment to security and trust, put IBM at the forefront of smarter health.


Learn more: