Desafíos y Problemas en la Extracción de Entidades Nombradas de las Notas Clínicas de Oncología

Autores/as

  • Luiz Henrique Pereira Niero Comsentimento NLP Lab
  • João Vitor Andrioli de Souza Comsentimento NLP Lab
  • Luciana Martins Gomes da Silva Comsentimento NLP Lab
  • Yohan Bonescki Gumiel Faculdade de Medicina da Universidade de São Paulo
  • Nícolas Henrique Borges Comsentimento NLP Lab
  • Gustavo Henrique Munhoz Piotto Comsentimento NLP Lab
  • Gustavo Giavarini Comsentimento NLP Lab
  • Lucas Emanuel Silva e Oliveira Comsentimento NLP Lab

DOI:

https://doi.org/10.59681/2175-4411.v15.iEspecial.2023.1097

Palabras clave:

Procesamiento de Lenguaje Natural, Registros Electrónicos de Salud, Oncología Médica

Resumen

Este artículo tiene como objetivo describir el proceso de anotación de un corpus multiinstitucional de textos clínicos en la especialidad de oncología y entrenar modelos para el Reconocimiento de Entidades Nombradas. Usamos el corpus anotado para entrenar modelos con diferentes cantidades de datos y comparamos el resultado del modelo con la cantidad de datos utilizados en el entrenamiento. El entrenamiento de los modelos se hizo a partir de la puesta a punto de las Representaciones de Codificadores Bidireccionales de Transformadores adaptados al dominio médico-biológico de la lengua portuguesa (BioBERTpt). Para comparar el comportamiento del modelo con el aumento de los datos de entrenamiento, los modelos se entrenaron con cantidades incrementales de datos. Como resultado, encontramos que los modelos entrenados con conjuntos de datos más pequeños pero completamente revisados funcionaron mejor que los modelos entrenados con conjuntos de datos más grandes con poca revisión.

Biografía del autor/a

Luiz Henrique Pereira Niero, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil. Paulista State University “Júlio de Mesquita Filho” - UNESP, Rio Claro (SP), Brasil.

João Vitor Andrioli de Souza, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil. Pontifical Catholic University of Paraná - PUCPR, Curitiba (PR), Brasil.

Luciana Martins Gomes da Silva, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil.

Yohan Bonescki Gumiel, Faculdade de Medicina da Universidade de São Paulo

Biomedical Informatics Laboratory - Instituto do Coração - HC FMUSP.

Nícolas Henrique Borges, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil.

Gustavo Henrique Munhoz Piotto, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil. DASA oncology, Brasil.

Gustavo Giavarini, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil.

Lucas Emanuel Silva e Oliveira, Comsentimento NLP Lab

Comsentimento NLP Lab, São Paulo, Brasil. Biomedical Informatics Laboratory - Instituto do Coração - HC FMUSP.

Citas

Jensen, PB; Jensen, LJ, Brunak, S. Mining electronic health records: towards better research applications and clinical care. Nature Reviews Genetics 2012;13(6): 395-405.

Jian F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al. Artificial intelligence in healthcare: Past, present and future, Stroke Vasc. Neurol. 2 2017;230–243.

Sheikhalishahi S, Miotto R, Dudley JT, Lavelli A, Rinaldi F, Osmani V. Natural language processing of clinical notes on chronic diseases: systematic review. JMIR medical informatics 2019;21:1-18.

World Health Organization. Cancer. Available from: https://www.who.int/news-room/fact-sheets/detail/cancer

Oliveira LES e, Peters AC, da Silva AMP, Gebeluca CP, Gumiel YB, Cintho LMM, et al. SemClinBr - a multi-institutional and multi-specialty semantically annotated corpus for Portuguese clinical NLP tasks. Journal of Biomedical Semantics. 2022 May 8;13(1).

Schneider ETR, de Souza JVA, Knafou J, Oliveira LES e, Copara J, Gumiel YB, et al. BioBERTpt - A Portuguese Neural Language Model for Clinical Named Entity Recognition. Proceedings of the 3rd Clinical Natural Language Processing Workshop. 2020;

de Souza JVA, Gumiel YB, Silva EL, Moro CM. Named entity recognition for clinical portuguese corpus with conditional random fields and semantic group.Anais do XIX Simpósio Brasileiro de Computação Aplicada à Saúde; 2019, 318-323.

dos Santos HDP, Silva AP, Maciel MCO, Burin HMV, Urbanetto JS, Vieira R. Fall Detection in EHR using Word Embeddings and Deep Learning. Proceedings of the 19th International Conference on Bioinformatics and Bioengineering (BIBE); 2019, 265-268.

Lopes F, Teixeira C, Oliveira HG. Contributions to Clinical Named Entity Recognition in Portuguese. Proceedings of the 18th BioNLP Workshop and Shared Task; 2019, 223–233.

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: A Cancer Journal for Clinicians [Internet]. 2021 Feb 4;71(3):209–49. Available from: https://acsjournals.onlinelibrary.wiley.com/doi/10.3322/caac.21660

Lample G, Ballesteros M, Subramanian S, Kawakami K, Dyer C. Neural Architectures for Named Entity Recognition [Internet]. arXiv.org. 2016. Available from: https://arxiv.org/abs/1603.01360

Roberts A, Gaizauskas R, Hepple M, Davis N, Demetriou G, Guo Y, et al. The CLEF corpus: semantic annotation of clinical text. AMIA Annual Symposium Proceedings AMIA Symposium [Internet]. 2007 [cited 2022 Aug 27];2007:625–9. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2655900/

Kang T, Zhang S, Tang Y, Hruby GW, Rusanov A, Elhadad N, et al. EliIE: An open-source information extraction system for clinical trial eligibility criteria. Journal of the American Medical Informatics Association [Internet]. 2017 Apr 1 [cited 2021 Dec 1];24(6):1062–71. Available from: https://academic.oup.com/jamia/article/24/6/1062/3098256?login=true

Richard LJ, Koch GG. The Measurement of Observer Agreement for Categorical Data. Biometrics [Internet]. 1977;33(1):159–74. Available from: https://www.jstor.org/stable/2529310?seq=7#metadata_info_tab_contents

Stubbs A. MAE and MAI: Lightweight Annotation and Adjudication Tools [Internet]. Association for Computational Linguistics; 2011 [cited 2022 Aug 29] p. 23–4. Available from: https://aclanthology.org/W11-0416.pdf

de Oliveira LFA, Oliveira LES, Gumiel YB, Carvalho DR, Moro CMC. Defining a state-of-the-art POS-tagging environment for Brazilian Portuguese clinical texts. Research on Biomedical Engineering. 2020 Jun 19;36(3):267–76.

Devlin J, Chang M-W, Lee K, Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding [Internet]. arXiv.org. 2018. Available from: https://arxiv.org/abs/1810.04805

CoNLL-2003 standard https://aclanthology.org/W03-0419.pdf

Souza JVA, Schneider ETR, Oliveira LES, Gumiel YB, Paraiso EC, Teodoro D, Barra CMCM. A Multilabel approach to Portuguese clinical named entity recognition. Journal of Health Informatics. 2020 Dez; (special number SBIS): 366-72.

Simpletransformers library. Available from: https://simpletransformers.ai/

Google Collab. Available from: https://colab.research.google.com/

Issifu AM, Ganiz MC. A simple data augmentation method to improve the performance of named entity recognition models in medical domain. 2021. 6th International Conference on Computer Science and Engineering (UBMK): 763-768.

Publicado

2023-07-20

Cómo citar

Niero, L. H. P., Souza, J. V. A. de, Silva, L. M. G. da, Gumiel, Y. B., Borges, N. H., Piotto, G. H. M., … Oliveira, L. E. S. e. (2023). Desafíos y Problemas en la Extracción de Entidades Nombradas de las Notas Clínicas de Oncología. Journal of Health Informatics, 15(Especial). https://doi.org/10.59681/2175-4411.v15.iEspecial.2023.1097

Artículos similares

1 2 3 4 5 6 7 8 9 10 > >> 

También puede {advancedSearchLink} para este artículo.

Artículos más leídos del mismo autor/a