EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

FHIR-Former: enhancing clinical predictions through Fast Healthcare Interoperability Resources and large language models

Merlin Engelke iD, Giulia Baldini iD, Jens Kleesiek iD, Felix Nensa iD, Amin Dada iD

Journal of the American Medical Informatics Association · 2025

DOI: 10.1093/jamia/ocaf165 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, 2025. Verfasst von Merlin Engelke und 4 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Artificial Intelligence in Medicine, University Medicine Essen, Essen, 45131, Germany. Hauptthemen laut MEDLINE: Health Information Interoperability, Electronic Health Records. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Open Access Publication Fund of the University of Duisburg-Essen.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

To address the challenges of data heterogeneity and manual feature engineering in clinical predictive modeling, we introduce FHIR-Former, an open-source framework integrating Fast Healthcare Interoperability Resources (FHIR) with large language models (LLMs) to automate and standardize clinical prediction tasks. Materials and Methods FHIR-Former dynamically processes structured (eg, lab results, medications) and unstructured (eg, clinical notes) data from FHIR resources. The pipeline supports multiple classification tasks, including 30-day readmission, imaging study prediction, and ICD code classification. Leveraging open-source LLMs (GeBERTa), we trained models on 1.1 million data points across ten FHIR resources using retrospective inpatient data (2018-2024). Hyperparameters were optimized via Bayesian methods, and outputs were mapped to FHIR RiskAssessment resources for interoperability.

Results

FHIR-Former achieved an F1-score of 70.7% and accuracy of 72.9% for 30-day readmission, 51.8% F1-score (88.1% accuracy) for mortality prediction, and 61% macro F1-score for imaging study classification. The ICD code prediction model attained 94% accuracy. Performance demonstrated promising performance for readmission and showed scalability across tasks without manual feature engineering.

Discussion

FHIR-Former eliminates institution-specific preprocessing by adapting to diverse FHIR implementations, enabling seamless integration of multimodal data. Its configurable architecture outperformed prior frameworks reliant on static inputs or limited to unstructured text. Real-time risk scores embedded in FHIR servers enhance clinical workflows without disrupting existing practices.

Conclusion

By harmonizing FHIR standardization with LLM flexibility, FHIR-Former advances scalable, interoperable predictive modeling in healthcare. The open-source framework facilitates automation, improves resource allocation, and supports personalized decision-making, bridging gaps between AI innovation and clinical practice.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Health Information InteroperabilityElectronic Health RecordsHumansBayes TheoremPatient ReadmissionRetrospective StudiesProgramming LanguagesMachine LearningNatural Language ProcessingLarge Language Models

Stichwörter der Autor:innen

Clinical Prediction TasksElectronic Health Records (EHR)Fast Healthcare Interoperability Resources (FHIR)Hospital ReadmissionICD Code ClassificationImaging Study PredictionIn-hospital MortalityLarge Language Models (LLMs)Multimodal Data

Autor:innen und Institutionen

  • Institute for Artificial Intelligence in Medicine, University Medicine Essen, Essen, 45131, Germany.
  • Institute of Diagnostic and Interventional Radiology and Neuroradiology, University Medicine Essen, Essen, 45147, Germany.
  • Faculty of Computer Science and Medical Faculty, University of Duisburg-Essen, Essen, 45127, Germany.
  • Department of Physics, TU Dortmund University, Dortmund, 44227, Germany.

ORCID-Profile: Merlin Engelke, Giulia Baldini, Jens Kleesiek, Felix Nensa, Amin Dada

Förderung

  • Open Access Publication Fund of the University of Duisburg-Essen

Zitieren

APA

Engelke, M., Baldini, G., Kleesiek, J., Nensa, F., & Dada, A. (2025). FHIR-Former: enhancing clinical predictions through Fast Healthcare Interoperability Resources and large language models. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocaf165

Vancouver

Engelke M, Baldini G, Kleesiek J, Nensa F, Dada A. FHIR-Former: enhancing clinical predictions through Fast Healthcare Interoperability Resources and large language models. Journal of the American Medical Informatics Association. 2025. doi:10.1093/jamia/ocaf165

BibTeX
@article{Engelke2025,
  author = {Engelke, Merlin and Baldini, Giulia and Kleesiek, Jens and Nensa, Felix and Dada, Amin},
  title = {FHIR-Former: enhancing clinical predictions through Fast Healthcare Interoperability Resources and large language models},
  journal = {Journal of the American Medical Informatics Association},
  year = {2025},
  doi = {10.1093/jamia/ocaf165}
}
Kostenlos starten →