EUVIMEDEUVIMED entdecken
Open Access

Multilingual Evidence-Based Question-Answering for Stroke Discharge Summaries: Study of Cross-Lingual Heterogeneity in Clinical Reports

Vojtěch Lanz, Aleksis Datseris, Svetla Boytcheva, Jiří Mayer, Šárka Zikánová, Robert Mikulík, Pavel Pecina

Journal of Medical Internet Research · 19.08.2026 · 28 · e96347-e96347

DOI: 10.2196/96347 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Medical Internet Research, online erschienen am 19.08.2026. Verfasst von Vojtěch Lanz und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Malostranské náměstí 25, Prague, 118 00, Czech R…. Hauptthemen laut MEDLINE: Stroke, Multilingualism, Language, Patient Discharge. Die Arbeit nennt 97 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The Registry of Stroke Care Quality (RES-Q) is a health care quality improvement platform used globally. RES-Q collects structured quality-of-care data for patients with stroke, requiring clinicians to manually extract information from electronic health records or documents such as discharge summaries. This process is essential but time-consuming, particularly given the variability, length, and semistructured nature of clinical reports.

Objective

This study aimed to develop and evaluate a multilingual Evidence-Based Question-Answering framework that identifies supporting text spans in clinical reports of patients with stroke and proposes answer suggestions for structured clinical forms, with the goal of reducing clinician workload while preserving full human oversight.

Methods

We conduct a multilingual study using more than 1500 pseudonymized stroke discharge summaries in 5 languages, annotated with question-evidence-answer triplets. Encoder-based language models are used to extract evidence spans from the reports, while generative language models are used to predict normalized form answers based on the extracted evidences. We compare multiple training strategies, such as (1) models trained on reports in a single target language, (2) models trained jointly on reports in different languages, and (3) models trained on original reports combined with cross-lingual data augmentations. We evaluate performance on Evidence Extraction, Answer Prediction, and end-to-end Evidence-Based Question Answering across the 5 languages.

Results

The presented Evidence-Based Question-Answering system achieves 88% end-to-end accuracy in form filling across 5 languages (77% for patient-specific questions and 95% for default or unverifiable items). Evidence Extraction is the primary bottleneck, reaching 85% F 1 and 79% exact match, whereas Answer Prediction based on extracted evidences is more stable, achieving 95% accuracy. The performance varies by question type, and cross-lingual training generally reduces Evidence Extraction performance but has little effect on Answer Prediction. Model performance is influenced more by reporting practices and dataset characteristics than by language itself.

Conclusions

Evidence-Based Question Answering over multilingual stroke discharge summaries enables human-in-the-loop validation and effective answer prediction with moderate computational resources. Evidence Extraction is the main bottleneck, while Answer Prediction is robust across languages and model sizes. The approach supports structured data collection, although generalization to new languages requires target-language training data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

StrokeMultilingualismLanguagePatient DischargePatient Discharge SummariesHumansElectronic Health Records

Stichwörter der Autor:innen

clinical question answeringevidence extractionmedical large language modelsmultilingual clinical natural language processingstroke discharge summaries

Autor:innen und Institutionen

  • Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Malostranské náměstí 25, Prague, 118 00, Czech Republic, 420 951 554 278.
  • Graphwise, Sofia, Bulgaria.
  • Computer Informatics, Faculty of Mathematics and Informatics, Sofia University "St. Kliment Ohridski", Sofia, Bulgaria.
  • Department of Linguistic Modelling and Knowledge Processing, Faculty of Mathematics and Informatics, Sofia University "St. Kliment Ohridski", Sofia, Bulgaria.
  • Health Management Institute, Brno, Czech Republic.

Zitieren

APA

Lanz, V., Datseris, A., Boytcheva, S., Mayer, J., Zikánová, Š., Mikulík, R., & Pecina, P. (2026). Multilingual Evidence-Based Question-Answering for Stroke Discharge Summaries: Study of Cross-Lingual Heterogeneity in Clinical Reports. Journal of Medical Internet Research, 28, e96347-e96347. https://doi.org/10.2196/96347

Vancouver

Lanz V, Datseris A, Boytcheva S, Mayer J, Zikánová Š, Mikulík R, et al. Multilingual Evidence-Based Question-Answering for Stroke Discharge Summaries: Study of Cross-Lingual Heterogeneity in Clinical Reports. Journal of Medical Internet Research. 2026;28:e96347-e96347. doi:10.2196/96347

BibTeX
@article{Lanz2026,
  author = {Lanz, Vojtěch and Datseris, Aleksis and Boytcheva, Svetla and Mayer, Jiří and Zikánová, Šárka and Mikulík, Robert and Pecina, Pavel},
  title = {Multilingual Evidence-Based Question-Answering for Stroke Discharge Summaries: Study of Cross-Lingual Heterogeneity in Clinical Reports},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e96347--e96347},
  doi = {10.2196/96347}
}
Kostenlos starten →