Real-World Data & Evidence

LynxCare rejoint le consortium OncoLab, un consortium public-privé de centres experts visant à faciliter et accélérer la recherche et l’innovation en oncologie. Le projet OncoLab a pour objectif de déployer des architectures de données avancées pour la recherche en oncologie au sein de trois institutions de référence dans le domaine : l’Institut Curie, l’Institut Bergonié et le Centre hospitalier universitaire de Toulouse.

LynxCare is proud to join the OncoLab consortium, a public-private consortium of expert centers to facilitate and accelerate research and innovation in oncology. The OncoLab project aims to deploy advanced data architectures for oncology research across three leading institutions in the field: Institut Curie, Institut Bergonié, and Toulouse University Hospital.

We are delighted to welcome Rhiannon Thomason as Chief Commercial Officer at LynxCare — joining us at a defining moment in our company’s journey. As we kick off the year, LynxCare continues to scale rapidly across Europe. Our federated network of hospitals is expanding, and an ever-growing ecosystem of partners is leveraging our platform to unlock high-quality, validated real-world health data that drives impactful research and supports better patient care.

Healthcare systems in Europe generate vast amounts of clinical information every day - from structured fields such as lab results to rich unstructured sources like physician notes, discharge summaries, and radiology reports. While structured data already supports a wide range of analytics and research, free-text documents contain an even deeper layer of clinical insight. Unlocking this potential requires advanced Natural Language Processing (NLP), yet general-purpose AI models are not optimally adapted to clinical language, multilingual environments, or the level of quality assurance needed for research and regulatory contexts.

As real-world evidence (RWE) increasingly informs clinical decision-making, regulatory assessments, and health-economics evaluations, the ability to generate insights from large, diverse, and quality-assured hospital datasets has become essential. LynxCare’s federated hospital network is designed to meet this need by combining EU geographic reach with a data quality framework and a robust governance framework, enabling multi-country, multi-site research while ensuring full compliance with privacy requirements.

80% of healthcare data is unstructured and underused. Barriers like lack of trust, site-specific complexity, and heavy IT needs keep it locked away. LynxCare’s Human-in-the-Loop + Auto-Retraining approach solves this with expert oversight, local configurability, and lightweight deployment—turning raw notes into reliable clinical insights.

Clinical concept normalization, the process of linking free-text mentions with clinical concepts from a standardized terminology, is essential for enabling analysis of healthcare data, especially Electronic Health Records (EHR), for research. However, the task presents significant challenges due to the extensive variation and ambiguity in medical text, especially in multilingual settings. In our technical report, we evaluated the LynxCare system for concept normalization and compared it against two popular open-source systems. Download full technical research paper by completing the form on the right.

In our study, we evaluated the performance of LynxCare’s clinical NLP pipeline against two open-source alternatives. Our goal was to assess out-of-the-box performance on multilingual biomedical text, particularly electronic health record (EHR) data.

Under the evolving European Health Data Space (EHDS) guidelines, the quality and reliability of electronic health record (EHR) data are critical for advancing AI-driven decision tools and clinical research. LynxCare introduces Sentinel — a robust, OMOP Common Data Model-based system designed to continuously measure, benchmark, and improve EHR-derived datasets for regulatory compliance, collaborative studies, and translational research.