Year: 2026
Type: preprint
Abstract: Background: Cardiovascular care generates large volumes of clinical free-text documentation in local languages. However, clinical natural language processing (NLP) resources remain concentrated in English, and no existing benchmark is cardiology-specific, multilingual, and broad in entity coverage. Methods: We present two contributions. The first is CardioCCC, a corpus of 508 cardiology case reports originally published in Spanish and extended to Czech, Dutch, English, Italian, Romanian, and Swedish through machine translation, annotation projection, and clinician review, with annotations covering diseases, symptoms, procedures, and medications. The second is CardioLM, comprising CardioBERTa, a set of language-specific encoders obtained by continued pretraining on cardiology and biomedical corpora, and CardioNER, a set of named entity recognition (NER) models fine-tuned on CardioCCC. Results: Cardiology-specific pretraining reduced pseudo-perplexity by up to a factor of eight, yet changed strict span-level F1 by at most 0.04, with benefits concentrated in languages whose backbones lacked biomedical coverage. The jointly trained multilingual XLM-RoBERTa-large models achieved the best strict F1 in nearly every language and entity type, exceeding the strongest monolingual system by up to 0.13 F1. A large share of residual errors concerned boundary detection: relaxed matching recovered 0.10-0.23 F1 for disease, procedure, and symptom, but only 0.01-0.07 for medication. Clinicians rated CardioNER well above a rule- and dictionary-based baseline, with the mean rating across entity types rising from 26% to 73%. Conclusions: Specialty-specific pretraining is not the decisive factor for downstream entity recognition, whereas multilingual joint training yields the strongest results in nearly all settings and a simpler deployment path for multi-country clinical information extraction.
Source: arXiv (Cornell University)
Authors: Danu, Manuela-Daniela, Rodríguez-Ortega, Miguel, Becerra-Tome, Alberto, Arancibia, Pablo, Gallego-Donoso, Fernando, Pratesi, Lorenzo, Bai, Xi, Dinescu, Andrei-Damian, Lima-López, Salvador, Guala, Andrea, Haitjema, Saskia, Mack, Jonas, Miranda, Michele, Nabki, Wesam Al, Rodríguez-Miret, Jan, Marica, Vasile George, Kincl, Vladimír, Dostálová, Kateřina, Andrej, Michael, Ujil, Alicia, Allaart, Corinne, Boonstra, Machteld, Overmars, Malin, Shah, Anoop, Ceroni, Giovanna, Kossoff, Jonathan, D'Amario, Domenico, Laborante, Renzo, Iaconelli, Antonio, Fumarulo, Isabella, Lenkowicz, Jacopo, Lilli, Livia, Weiss, Emma, Scafa, Alex, Lindberg, Felix, Gamage, Milou, Farré-Maduell, Eulàlia, Bratières, Sébastien, Krallinger, Martin, van Es, Bram
Institutions: Transilvania University of Brașov, Siemens (Romania), Barcelona Supercomputing Center, University of Amsterdam, Karolinska Institutet, University College London, Instituto de Salud Carlos III, Centro de Investigación en Red en Enfermedades Cardiovasculares, Vall d'Hebron Institut de Recerca, University Medical Center Utrecht, Masaryk University, Sapienza University of Rome, International Clinical Research Center, St. Anne's University Hospital Brno, St. Anne's University Hospital Brno, University Hospital Hradec Králové, Amsterdam UMC Location University of Amsterdam, Amsterdam University Medical Centers, Amsterdam Cardiovascular Sciences, Agostino Gemelli University Polyclinic, Istituti di Ricovero e Cura a Carattere Scientifico, Clinical Emergency Hospital Bucharest, Stockholm South General Hospital
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7225375533