NLP2FHIR: A FHIR-based Clinical Data Normalization Pipeline

The next generation phenotyping of Electronic Health Record (EHR) features the identification of true patient state in an accurate and high-throughput manner. To realize this vision, there is an urgent need to improve the reproducibility and interpretability of the underlying phenotype models and algorithms through a standards-based framework. The Fast Healthcare Interoperability Resources (FHIR) standard was developed to meet a variety of clinical interoperability needs. We developed a FHIR-based clinical data normalization pipeline known as NLP2FHIR at the Mayo Clinic. The core of the pipeline is a FHIR-based common type system that is used to harmonize and standardize the outputs from a number of clinical natural language processing (NLP) tools such as cTAKES, MedXN and MedTime.  In the context of secondary use of EHR data, we envision that a FHIR standard-driven data normalization pipeline would improve semantic interoperability between heterogeneous resources and tools, and enable effective exchange, integration, sharing and reuse of encoded and structured clinical narratives, along with well structured EHR data.