Alberto Marfoglia

Background: Given the exponential increase in clinical data, which accounts for around 30% of global data volume, effective information management has become crucial to ensuring robust interoperability. This trend is further expedited by implementing consumer-oriented Internet of Things platforms, contributing to the growth of the $8.3 trillion healthcare industry. These advancements, combined with challenges such as heterogeneous data formats and a lack of incentives, necessitate the development of pragmatic infrastructures and tools that harness contemporary clinical standards like Fast Healthcare Interoperability Resources (FHIR). Objective: This study aims to present a modular conversion pipeline employing a templating strategy for translating clinical data into the FHIR model. Emphasis is placed on utilizing a standard mapping specification like FHIR Mapping Language. This ensures essential properties such as platform independence, portability, and code reusability. Methods: The pipeline was developed incrementally, dividing its core functionalities into five modules: Input, Refinement, Mapping, Validation, and Export. These were subsequently validated by converting a dataset from a prosthetic fitting and rehabilitation center to demonstrate the approach’s validity in a real-world data context. Results: A total of 1962 hospital stay records of 1006 unique patients were converted successfully to 15 distinct types of FHIR resources. The successful conversion states the pipeline’s effectiveness, additionally showcasing its capabilities for enhancing semantic interoperability and facilitating the reuse of real-world data. Conclusion: Our approach emerges as a modular data conversion framework that addresses the limitations of existing solutions, making significant contributions to the creation of standardized, interoperable, and high-quality clinical datasets that serve as a foundation for further work.

Objective To explore the adoption and practical implementation of the three major health data standards (i.e., FHIR, OMOP-CDM, and openEHR), to evaluate their maturity level in terms of how extensively they have been applied and integrated into everyday clinical and research practice. Methods We conducted a systematic review registered in PROSPERO (CRD42024623398) following PRISMA guidelines. Literature searches were performed through PubMed, Cochrane, Scopus, Web of Science, and IEEE Xplore from 2021 to 2024. After de-duplication and screening, 99 studies were included. Data was extracted and classified according to five health application domains and five use cases based on the intended purpose of the standard in the work. Studies were assessed for implementation scale, ETL tools, coverage of the standard (i.e., the number of mapped source variables), and whether standards were adapted or used as-is. Results Of the 99 included studies, 57% used OMOP-CDM, 39% FHIR, and 8% openEHR. Most applications occurred in research settings (87%) and focused on data reuse (47%) or clinical decision support (23%). OMOP-CDM was preferred for large-scale, longitudinal research, while FHIR was dominant in the public health domain and for real-time data exchange. Only 27% of studies reported the coverage of the standard. FHIR implementations often require customization, complicating interoperability. OMOP-CDM offered strong analytical tooling but posed challenges for mapping and data loss. Few studies using openEHR reported limitations, with its uptake remaining limited. Conclusion Although FHIR, OMOP-CDM, and openEHR hold significant potential to enhance interoperability, their adoption remains fragmented. Each standard shows specific strengths: FHIR for exchange, OMOP-CDM for analytics, and openEHR for data persistence. A hybrid approach and clearer implementation practices are essential to support scalable, interoperable health data ecosystems.

The application of machine learning on healthcare data is often hindered by the lack of standardized and semantically explicit representation, leading to limited interoperability and reproducibility across datasets and experiments. The Medical Event Data Standard (MEDS) addresses these issues by introducing a minimal, event-centric data model designed for reproducible machine-learning workflows from health data. However, MEDS is defined as a data-format specification and does not natively provide integration with the Semantic Web ecosystem. In this article, we introduce MEDS-OWL, a lightweight OWL ontology that provides formal concepts and relations to represent MEDS datasets as RDF graphs. Additionally, we implemented meds2rdf, a Python conversion library that transforms MEDS events into RDF graphs, ensuring conformance with the ontology. We evaluate the proposed approach on two datasets: a synthetic clinical cohort describing care pathways for ruptured intracranial aneurysms, and a real-world subset of MIMIC-IV. To assess semantic consistency, we performed a SHACL validation against the resulting knowledge graphs. The first release of MEDS-OWL comprises 13 classes, 10 object properties, 20 data properties, and 24 OWL axioms. Combined with meds2rdf, it enables data transformation into FAIR-aligned datasets, provenance-aware publishing, and interoperability of event-based clinical data. By bridging MEDS with the Semantic Web, this work contributes a reusable semantic layer for event-based clinical data and establishes a robust foundation for subsequent graph-based analytics.

open

Background: The healthcare sector faces diverse challenges, including poor interoperability and a lack of personalized approaches, which limit patient outcomes. Ineffective data exchange and one-size-fits-all treatments fail to meet individual needs. Emerging technologies like digital twins (DTs), the semantic web, and AI show promise in tackling these obstacles. For this reason, we introduced CONNECTED, a conceptual multi-level framework that combines these techniques to deploy general-purpose patient DTs. Objective: This study assesses CONNECTED’s comprehensiveness, applicability, and utility for developing intelligent, personalized healthcare applications. Specifically, we deliver a preliminary version of the framework to predict future patient states and demonstrate its automation benefits in deploying semantically enriched, AI-powered patient DTs. Methods: We enhanced the CONNECTED architecture by providing a formal definition of DT and modularizing its core functionalities into four microservices (Properties, State, Capabilities, and Manifest). The Manifest service facilitates AI model integration through the Model Interface Manifest Ontology (MIMO), enabling automatic data-to-model binding via a reasoner. Using the HeartBeatKG quality assessment tool, we validated MIMO and tested the internal logic by integrating a well-established stroke-risk model. Results: Our implementation comprehends: (1) deploying a FHIR-compliant, patient-centric API for clinical history access, real-time monitoring, and predictive simulation; (2) publishing MIMO; (3) establishing the Manifest protocol for seamless, general-purpose AI model integration tailored to individual patient profiles; and (4) a proof-of-concept benchmarking application comparing multiple stroke risk classifiers. Conclusion: CONNECTED establishes a flexible, scalable foundation for interoperable semantic patient DTs. Automation reduces technical overhead and enables users to focus on delivering personalized, insight-driven care.