Cross-lingual clinical annotation improves medical text tagging accuracy
Cross-Lingual Clinical Annotation Projection as Constrained Text Generation: A Six-Language Study
Computation and LanguageArtificial Intelligence
Summary
Labeling medical text in many languages is hard because each language needs special tools and data. The authors tested a new way to use large language models (LLMs) to add medical labels directly into text in six different languages. This approach worked better than older methods that picked label spans in the text. Their method helps make accurate and reusable medical datasets across languages with less expert work and cost.
What this means in practice
- •For hospital data teams: Create accurate multilingual clinical text annotations to improve medical data analysis and machine learning models in multiple languages simultaneously.
- •For healthcare software developers: Integrate direct LLM-based entity tagging to build clinical NLP tools supporting languages with limited existing annotated data and language-specific resources.
Authors
Álvaro Rey-Blanes, Francisco J. Moreno-Barea, Francisco J. Veredas
Abstract
Background: To determine whether cross-lingual clinical annotation projection can be formulated as a text-preserving, document-level generative task that produces verifiable character-level annotations for multilingual clinical corpus construction, and to characterize its robustness and computational trade-offs relative to candidate-based projection pipelines. Methods: We developed a constrained LLM projection workflow that inserts entity tags directly into immutable target-language text, followed by deterministic validation and character-offset reconstruction. We evaluated it alongside supervised candidate-span projection and hybrid ML-LLM refinement for transferring Spanish Disease, Symptom, and Procedure annotations into six languages. Evaluation used MultiClinAI gold standard with strict span matching and character-overlap F1 Results: Direct LLM projection achieved the strongest and most consistent performance. GLM 5.2 obtained a mean Strict F1 of 0.9201 across 18 language-entity combinations, while locally deployable Gemma4:31B achieved 0.9133. The best LLM configuration improved Strict F1 over the previous state of the art in all 18 settings, by 0.0564-0.1512, yielding 55,416 grounded mentions with reconstructed offsets. Conclusions: Direct LLM-based projection enables high-quality multilingual clinical annotation transfer and provides a practical approach for extending clinical NLP resources to languages with fewer annotated datasets and language-specific tools. Combined with local inference and deterministic validation, it can substantially reduce expert time and cost for multilingual clinical corpus construction.