Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Dataset for Pharmacovigilance in German, French, and Japanese: Annotating Adverse Drug Reactions across Languages

Lisa Raithel, Hui-Syuan Yeh, Shuntaro Yada, Cyril Grouin, Thomas Lavergne, Aurélie Névéol, Patrick Paroubek, Philippe Thomas, Tomohiro Nishiyama, Sebastian Möller, Eiji Aramaki, Yuji Matsumoto, Roland Roller, Pierre Zweigenbaum

TL;DR

This work presents a multilingual corpus of texts concerning ADRs gathered from diverse sources, including patient fora, social media, and clinical reports in German, French, and Japanese, which contributes to the development of real-world multilingual language models for healthcare.

Abstract

User-generated data sources have gained significance in uncovering Adverse Drug Reactions (ADRs), with an increasing number of discussions occurring in the digital world. However, the existing clinical corpora predominantly revolve around scientific articles in English. This work presents a multilingual corpus of texts concerning ADRs gathered from diverse sources, including patient fora, social media, and clinical reports in German, French, and Japanese. Our corpus contains annotations covering 12 entity types, four attribute types, and 13 relation types. It contributes to the development of real-world multilingual language models for healthcare. We provide statistics to highlight certain challenges associated with the corpus and conduct preliminary experiments resulting in strong baselines for extracting entities and relations between these entities, both within and across languages.

A Dataset for Pharmacovigilance in German, French, and Japanese: Annotating Adverse Drug Reactions across Languages

TL;DR

This work presents a multilingual corpus of texts concerning ADRs gathered from diverse sources, including patient fora, social media, and clinical reports in German, French, and Japanese, which contributes to the development of real-world multilingual language models for healthcare.

Abstract

User-generated data sources have gained significance in uncovering Adverse Drug Reactions (ADRs), with an increasing number of discussions occurring in the digital world. However, the existing clinical corpora predominantly revolve around scientific articles in English. This work presents a multilingual corpus of texts concerning ADRs gathered from diverse sources, including patient fora, social media, and clinical reports in German, French, and Japanese. Our corpus contains annotations covering 12 entity types, four attribute types, and 13 relation types. It contributes to the development of real-world multilingual language models for healthcare. We provide statistics to highlight certain challenges associated with the corpus and conduct preliminary experiments resulting in strong baselines for extracting entities and relations between these entities, both within and across languages.
Paper Structure (52 sections, 7 figures, 9 tables)

This paper contains 52 sections, 7 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methods
  4. Existing Datasets
  5. Spanish
  6. Russian
  7. French
  8. Japanese
  9. German
  10. The Corpus
  11. Data Collection
  12. German
  13. French
  14. Japanese
  15. Annotation
  16. ...and 37 more sections

Figures (7)

  • Figure 1: Example annotation of a German (top) and Japanese (bottom) text, with their respective English translation.
  • Figure 2: An example annotation of a caused relation taken from the French dataset and translated. According to the writer of this message (patient), the medication infliximab is likely to have caused the symptoms toute rouge and nauséeuse.
  • Figure 3: The distribution of document length of the German (a), French (b), and Japanese (c) data using the number of tokens. Note the different scaling on the axes.
  • Figure 4: The distribution of entity types across all documents for German (a), French (b), and Japanese (ja). Note the difference in scale when comparing the three languages.
  • Figure 5: The distribution of span length per entity type for German (a), French (b), and Japanese (c).
  • ...and 2 more figures