FHIR Transform Engine

‍

To integrate health, social care and clinical research applications using FHIR you need to build and maintain transforms between local or legacy data formats and FHIR. Our FHIR Transform Engine (FTE) is the most cost-effective way to do this.

‍

Using FTE, you rarely need to write code. Instead, you use a graphical Mapping Editor to make declarative mappings between your other data formats and FHIR. What are the benefits of this approach?

• Faster: Mapping is a skill; but once you have it, you can develop transforms much more quickly than by writing code
• Two-way transforms and round trips: One set of mappings gives transforms in both directions; FHIR=>other and other=>FHIR
• More Accurate: Using mature automated test tools, you can immediately test any change to mappings in a rapid map/test/map/test cycle
• Transparent and maintainable: Mappings are easy to understand, easy to review and maintain. Key design decisions are not buried in code.
• Supports any set of FHIR Profiles: Can transform to bundles and resources in any set of FHIR profiles such as the FHIR UK Core profiles
• Escape to code when needed: There are aspects of transforms (such as finding SNOMED codes) where procedural code may be simplest. You can easily build these in to mapping-based transforms, with a high level of reuse

‍

There are numerous use cases for FTE including:

• Build a RESTful FHIR Server on any existing healthcare application.
• Incorporate transforms generated by FTE into middleware, e.g. integration engines
• Batch Population of FHIR Server from a non-FHIR source
• Batch Extraction from FHIR Server into a non-FHIR source
• As a preparatory step for using our Cross Query data quality auditing tool
• As a preparatory step for using our Smart Search demographics matching and de-duping tool

‍

For more in-depth information, take a look at our Technology page.

‍

Accelerate your HL7v2 to/from FHIR mapping project

‍

HL7 International has a project to develop standard mappings from HL7 Version 2 to FHIR. The mappings are defined in spreadsheets. They already address the commonly used V2 messages, and all widely used V2 segments. They embody the results of many years’ V2 and FHIR expertise. The spreadsheets are typically used manually to develop V2-FHIR transforms in other tools or programming languages.

FTE can capture the spreadsheets automatically and convert them to FTE mappings. Without any modification, these FTE mappings deliver powerful standard V2 to FHIR transforms, capturing the great majority of the information in the V2 messages.

The FTE mappings are a tremendous starting point for developing working local transforms, from local variants of Version 2 to local profiles of FHIR. Localisation is always needed. To do it, you only need to edit the FTE mappings to reflect local features, rather than starting from scratch.

The V2-to-FHIR transforms are delivered as generated Java, and run very quickly. A typical transform is many thousands of lines of code, reflecting the huge effort that would be needed to develop it by hand coding. It can be easily maintained or adapted, just by editing the FTE mappings and re-generating.

‍

‍