A Standardized Clinical Data Harmonization Pipeline for Scalable AI Application Deployment (FHIR-DHP): Validation and Usability Study (Preprint)

Abstract

BACKGROUND Increasing digitalisation in the medical domain gives rise to large amounts of healthcare data which has the potential to expand clinical knowledge and transform patient care if leveraged through artificial intelligence (AI). Yet, big data and AI oftentimes cannot unlock their full potential at scale, owing to non-standardised data formats, lack of technical and semantic data interoperability, and limited cooperation between stakeholders in the healthcare system. Despite the existence of standardised data formats for the medical domain, such as Fast Healthcare Interoperability Resources (FHIR), their prevalence and usability for AI remains limited. OBJECTIVE We developed a data harmonisation pipeline (DHP) for clinical data sets relying on the common FHIR data standard. METHODS We validated the performance and usability of our FHIR-DHP with data from the MIMIC IV database including > 40,000 patients admitted to an intensive care unit. RESULTS We present the FHIR-DHP workflow in respect of transformation of “raw” hospital records into a harmonised, AI-friendly data representation. The pipeline consists of five key preprocessing steps: querying of data from hospital database, FHIR mapping, syntactic validation, transfer of harmonised data into the patient-model database and export of data in an AI-friendly format for further medical applications. A detailed example of FHIR-DHP execution was presented for clinical diagnoses records. CONCLUSIONS Our approach enables scalable and needs-driven data modelling of large and heterogenous clinical data sets. The FHIR-DHP is a pivotal step towards increasing cooperation, interoperability and quality of patient care in the clinical routine and for medical research. CLINICALTRIAL Data interoperability, FHIR, data standardisation pipeline, MIMIC IV