EUVIMEDScopri EUVIMED
Open access

Deep Learning Transformer Models for Building a Comprehensive and Real-time Trauma Observatory: Development and Validation Study

Gabrielle Chenais iD, Cédric Gil-Jardiné iD, Hélène Touchais iD, Marta Avalos Fernandez iD, Benjamin Contrand iD, Eric Tellier iD, Xavier Combes iD, Loick Bourdois iD, Philippe Revel iD, Emmanuel Lagarde iD

JMIR AI · 2023-01-12 · 2 · e40843

DOI: 10.2196/40843 ↗

At a glance

Journal article in JMIR AI, published online on 2023-01-12. Written by Gabrielle Chenais and 9 co-authors. Institution: Unit 1219, Bordeaux Public Health Center, Institut National de la Santé et de la Recherche Médicale, Bordeaux, France. Cited 13 times so far (Crossref); 48 references. The full text is freely available in PubMed Central.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Background

Public health surveillance relies on the collection of data, often in near-real time. Recent advances in natural language processing make it possible to envisage an automated system for extracting information from electronic health records.

Objective

To study the feasibility of setting up a national trauma observatory in France, we compared the performance of several automatic language processing methods in a multiclass classification task of unstructured clinical notes.

Methods

A total of 69,110 free-text clinical notes related to visits to the emergency departments of the University Hospital of Bordeaux, France, between 2012 and 2019 were manually annotated. Among these clinical notes, 32.5% (22,481/69,110) were traumas. We trained 4 transformer models (deep learning models that encompass attention mechanism) and compared them with the term frequency–inverse document frequency associated with the support vector machine method.

Results

The transformer models consistently performed better than the term frequency–inverse document frequency and a support vector machine. Among the transformers, the GPTanam model pretrained with a French corpus with an additional autosupervised learning step on 306,368 unlabeled clinical notes showed the best performance with a micro F1-score of 0.969.

Conclusions

The transformers proved efficient at the multiclass classification of narrative and medical data. Further steps for improvement should focus on the expansion of abbreviations and multioutput multiclass classification.

Subjects

Author keywords

deep learningemergenciesnatural language processingpublic healthtransformerstrauma

Authors and institutions

  • Unit 1219, Bordeaux Public Health Center, Institut National de la Santé et de la Recherche Médicale, Bordeaux, France.
  • Emergency Department, Bordeaux University Hospital, Bordeaux, France.
  • Statistics in Systems Biology and Translational Medicine Team, University of Bordeaux, Institut National de Recherche en Sciences et Technologies du Numérique, Talence, France.

ORCID profiles: Gabrielle Chenais, Cédric Gil-Jardiné, Hélène Touchais, Marta Avalos Fernandez, Benjamin Contrand, Eric Tellier, Xavier Combes, Loick Bourdois, Philippe Revel, Emmanuel Lagarde

Cite

APA

Chenais, G., Gil-Jardiné, C., Touchais, H., Avalos Fernandez, M., Contrand, B., Tellier, E., Combes, X., Bourdois, L., Revel, P., & Lagarde, E. (2023). Deep Learning Transformer Models for Building a Comprehensive and Real-time Trauma Observatory: Development and Validation Study. JMIR AI, 2, e40843. https://doi.org/10.2196/40843

Vancouver

Chenais G, Gil-Jardiné C, Touchais H, Avalos Fernandez M, Contrand B, Tellier E, et al. Deep Learning Transformer Models for Building a Comprehensive and Real-time Trauma Observatory: Development and Validation Study. JMIR AI. 2023;2:e40843. doi:10.2196/40843

BibTeX
@article{Chenais2023,
  author = {Chenais, Gabrielle and Gil-Jardiné, Cédric and Touchais, Hélène and Avalos Fernandez, Marta and Contrand, Benjamin and Tellier, Eric and Combes, Xavier and Bourdois, Loick and Revel, Philippe and Lagarde, Emmanuel},
  title = {Deep Learning Transformer Models for Building a Comprehensive and Real-time Trauma Observatory: Development and Validation Study},
  journal = {JMIR AI},
  year = {2023},
  volume = {2},
  pages = {e40843},
  doi = {10.2196/40843}
}
Start for free →