EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Fake It till You Predict It: Data Augmentation Strategies to Detect Initiation and Termination of Oncology Treatment

Valentin Pohyer, Elizabeth Fabre, Stéphane Oudard, Laure Fournier, Bastien Rance

Studies in Health Technology and Informatics · 2025

DOI: 10.3233/shti250468 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, 2025. Verfasst von Valentin Pohyer und 4 weiteren Autor:innen. Beteiligte Einrichtung: Hôpital Européen Georges Pompidou, AP-HP, Paris, France. Hauptthemen laut MEDLINE: Natural Language Processing, Neoplasms, Antineoplastic Agents, Electronic Health Records. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

At the hospital, the dispersion of information regarding anti-cancer treatment makes it difficult to extract. We proposed a solution capable of identifying dates, drugs and their temporal relationship within free-text oncology reports with very few manual annotations. We used pattern recognition for dates, dictionaries for drugs and transformer language models for the relationship, combined with a data augmentation strategy. Our models achieved good prediction F1-scores, reaching 0.872. The performance of models with data augmentation outperforms those of models without. By inferring such models, we can now identify and structure thousands of previously unavailable treatment events to better apprehend solutions and patient response.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Natural Language ProcessingNeoplasmsAntineoplastic AgentsElectronic Health RecordsData MiningHumansMedical Oncology

Stichwörter der Autor:innen

BERTCancerData augmentationFrenchHealthNLPTransformersTreatment

Substanzen

Antineoplastic Agents

Autor:innen und Institutionen

  • Hôpital Européen Georges Pompidou, AP-HP, Paris, France.
  • Inserm, Centre de Recherche des Cordeliers, Université Sorbonne-Paris Cité, Paris, France.
  • Université Paris Cité, Paris, France.
  • Inserm, PARCC, Paris, France.

Zitieren

APA

Pohyer, V., Fabre, E., Oudard, S., Fournier, L., & Rance, B. (2025). Fake It till You Predict It: Data Augmentation Strategies to Detect Initiation and Termination of Oncology Treatment. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti250468

Vancouver

Pohyer V, Fabre E, Oudard S, Fournier L, Rance B. Fake It till You Predict It: Data Augmentation Strategies to Detect Initiation and Termination of Oncology Treatment. Studies in Health Technology and Informatics. 2025. doi:10.3233/shti250468

BibTeX
@article{Pohyer2025,
  author = {Pohyer, Valentin and Fabre, Elizabeth and Oudard, Stéphane and Fournier, Laure and Rance, Bastien},
  title = {Fake It till You Predict It: Data Augmentation Strategies to Detect Initiation and Termination of Oncology Treatment},
  journal = {Studies in Health Technology and Informatics},
  year = {2025},
  doi = {10.3233/shti250468}
}
Kostenlos starten →