EUVIMEDEUVIMED entdecken
Open Access

Impact of Translation on Biomedical Information Extraction: Experiment on Real-Life Clinical Notes

Christel Gérardin, Yuhan Xiong, Perceval Wajsbürt, Fabrice Carrat, Xavier Tannier

JMIR Medical Informatics · 04.04.2024 · 12 · e49607-e49607

DOI: 10.2196/49607 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JMIR Medical Informatics, online erschienen am 04.04.2024. Verfasst von Christel Gérardin und 4 weiteren Autor:innen. Beteiligte Einrichtung: Institut Pierre Louis d'Epidémiologie et de Santé Publique, Sorbonne Université, Institut National de la Santé et de la Recherche Médicale, Paris, Fr…. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Biomedical natural language processing tasks are best performed with English models, and translation tools have undergone major improvements. On the other hand, building annotated biomedical data sets remains a challenge.

Objective

The aim of our study is to determine whether the use of English tools to extract and normalize French medical concepts based on translations provides comparable performance to that of French models trained on a set of annotated French clinical notes.

Methods

We compared 2 methods: 1 involving French-language models and 1 involving English-language models. For the native French method, the named entity recognition and normalization steps were performed separately. For the translated English method, after the first translation step, we compared a 2-step method and a terminology-oriented method that performs extraction and normalization at the same time. We used French, English, and bilingual annotated data sets to evaluate all stages (named entity recognition, normalization, and translation) of our algorithms.

Results

The native French method outperformed the translated English method, with an overall F 1 -score of 0.51 (95% CI 0.47-0.55), compared with 0.39 (95% CI 0.34-0.44) and 0.38 (95% CI 0.36-0.40) for the 2 English methods tested.

Conclusions

Despite recent improvements in translation models, there is a significant difference in performance between the 2 approaches in favor of the native French method, which is more effective on French medical texts, even with few annotated documents.

Schlagwörter

Stichwörter der Autor:innen

bilingual language modelbiomedical data setconcept normalizationnamed entity recognitionnatural language processingtranslationtranslational tool

Autor:innen und Institutionen

  • Institut Pierre Louis d'Epidémiologie et de Santé Publique, Sorbonne Université, Institut National de la Santé et de la Recherche Médicale, Paris, France.
  • Shanghai Jiaotong University, Shanghai, China.
  • Innovation and Data Unit, Assistance Publique Hôpitaux de Paris, Paris, France.
  • Department of Public Health, Assistance Publique Hôpitaux de Paris, Hôpital Saint-Antoine, Paris, France.
  • 5, Sorbonne Université, Institut National de la Santé et de la Recherche Médicale, Université Sorbonne Paris-Nord, Laboratoire d'Informatique Médicale et de Connaissance en e-Santé, Paris, France.

Zitieren

APA

Gérardin, C., Xiong, Y., Wajsbürt, P., Carrat, F., & Tannier, X. (2024). Impact of Translation on Biomedical Information Extraction: Experiment on Real-Life Clinical Notes. JMIR Medical Informatics, 12, e49607-e49607. https://doi.org/10.2196/49607

Vancouver

Gérardin C, Xiong Y, Wajsbürt P, Carrat F, Tannier X. Impact of Translation on Biomedical Information Extraction: Experiment on Real-Life Clinical Notes. JMIR Medical Informatics. 2024;12:e49607-e49607. doi:10.2196/49607

BibTeX
@article{Grardin2024,
  author = {Gérardin, Christel and Xiong, Yuhan and Wajsbürt, Perceval and Carrat, Fabrice and Tannier, Xavier},
  title = {Impact of Translation on Biomedical Information Extraction: Experiment on Real-Life Clinical Notes},
  journal = {JMIR Medical Informatics},
  year = {2024},
  volume = {12},
  pages = {e49607--e49607},
  doi = {10.2196/49607}
}
Kostenlos starten →