Comparando enfoques deep learning en una fase y en dos fases para extraer interacciones farmacológicas de texto

• Autores: Antonio Miranda Escalada, Isabel Segura Bedmar
• Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 64, 2020, págs. 69-76
• Idioma: español
• Títulos paralelos:
  • One stage versus two stages deep learning approaches for the extraction of drug-drug interactions from texts
• Enlaces
  • Texto completo
• Dialnet Métricas: 1 Cita
• Resumen
  • español

    Las interacciones farmacológicas (DDI) son una de las causas de reacciones adversas a medicamentos. Ocurren cuando una medicina interfiere en la acción de una segunda. En la actualidad, no existe una base de datos completa y actualizada donde los profesionales de la salud puedan consultar las interacciones de cualquier medicamento porque la mayor parte del conocimiento sobre DDIs está oculto en texto no estructurado. En los últimos años, el aprendizaje profundo se ha aplicado con éxito a la extracción de DDIs de los textos, lo que requiere la detección y posterior clasificación de DDIs. La mayoría de los sistemas de aprendizaje profundo para extracción de DDIs desarrollados hasta ahora han abordado la detección y clasificación en un solo paso. En este estudio, comparamos el rendimiento de las arquitecturas de una y dos etapas para la extracción de DDI. Nuestras arquitecturas se basan en una capa de red neuronal recurrente bidireccional compuesta de Gated Recurrent Units (GRU). El sistema en dos etapas obtuvo un puntaje F1 promedio de 67.45 % en el dataset de evaluación.

  • English

    Drug-drug interactions (DDI) are a cause of adverse drug reactions. They occur when a drug has an impact on the effect of another drug. There is not a complete, up to date database where health care professionals can consult the interactions of any drug because most of the knowledge on DDI is hidden in unstructured text. In last years, deep learning has been succesfully applied to the extraction of DDI from texts, which requires the detection and later classification of DDI. Most of the deep learning systems for DDI extraction developed so far have addressed the detection and classification in one single step. In this study, we compare the performance of one-stage and two-stage architectures for DDI extraction. Our architectures are based on a bidirectional recurrent neural network layer composed of Gated Recurrent Units. The two-stage system obtained a 67.45 % micro-average F1 score on the test set.

• Referencias bibliográficas
  • Abacha, A. B., M. F. M. Chowdhury, A. Karanasiou, Y. Mrabet, A. Lavelli, and P. Zweigenbaum. 2015. Text mining for pharmacovigilance: Using...
  • Cho, K., B. Van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio. 2014. Learning phrase representations using...
  • Chowdhury, M. F. M. and A. Lavelli. 2013. Exploiting the scope of negations and heterogeneous features for relation extraction: A case study...
  • Chung, J., C. Gulcehre, K. Cho, and Y. Bengio. 2014. Empirical evaluation of gated recurrent neural networks on sequence modeling. In NIPS...
  • Dewi, I. N., S. Dong, and J. Hu. 2017. Drugdrug interaction relation extraction with deep convolutional neural networks. In IEEE International...
  • Glorot, X. and Y. Bengio. 2010. Understanding the difficulty of training deep feedforward neural networks. In 13th International Conference...
  • Herrero-Zazo, M., I. Segura-Bedmar, P. Martínez, and T. Declerck. 2013. The DDI corpus: An annotated corpus with pharmacological substancesand...
  • Kavuluru, R., A. Rios, and T. Tran. 2017. Extracting Drug-Drug interactions with Word and Character-Level Recurrent Neural Networks. In IEEE...
  • Kim, S., H. Liu, L. Yeganova, and W. J. Wilbur. 2015. Extracting drug– drug interactions from literature using a rich feature-based linear...
  • Lai, S., L. Xu, K. Liu, and J. Zhao. 2015. Recurrent convolutional neural networks for text classification. In Twenty-ninth AAAI conference...
  • McNemar, Q. 1947. Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika, 12.
  • Mohamed, A.-R., G. Hinton, and G. Penn. 2010. Understanding the difficulty of training deep feedforward neural networks. In Acoustics, Speech...
  • Pyysalo, S., F. Ginter, H. Moen, T. Salakoski, and S. Ananiadou. 2013. Distributional Semantics Resources for Biomedical Text Processing....
  • Sainath, T. N., O. Vinyals, A. Senior, and H. Sak. 2015. Convolutional, long shortterm memory, fully connected deep neural networks. In Acoustics,...
  • Segura-Bedmar, I. 2010. Application of Information Extraction techniques to pharmacological domain: Extracting drug-drug interactions. Ph.D....
  • Segura-Bedmar, I., P. Martinez, and C. de Pablo-Sánchez. 2011. Using a shallow linguistic kernel for drug– drug interaction extraction. Journal...
  • Segura-Bedmar, I., P. Martinez, and M. Herrero-Zazo. 2013. SemEval-2013 task 9 : Extraction of drug-drug interactions from biomedical texts...
  • Segura Bedmar, I., P. Martinez, and D. Sánchez Cisneros. 2011. The 1st ddiextraction-2011 challenge task: Extraction of drug-drug interactions...
  • Suárez-Paniagua, V. and I. Segura-Bedmar. 2018. Evaluation of pooling operations in convolutional architectures for drug-drug interaction...
  • Sun, X., L. Ma, X. Du, J. Feng, and K. Dong. 2018. Deep Convolution Neural Networks for Drug-Drug Interaction Extraction. In IEEE International...
  • Thomas, P., M. Neves, T. Rocktäschel, and U. Leser. 2013. Wbi-ddi: drug-drug interaction extraction using majority voting. In Second Joint...
  • Wishart, D. e. a. 2017. Drugbank 5.0: a major update to the drugbank database for 2018. Nucleid Acird Research, 8.
  • Yi, Z., S. Li, J. Yu, and Q. Wu. 2017. Drugdrug Interaction Extraction via Recurrent Neural Network with Multiple Attention Layers. In International...
  • Zaremba, W., I. Sutskever, and O. Vinyals. 2015. Recurrent Neural Network Regularization. arXiv preprint arXiv:1409.2329, 1.
  • Zheng, W., H. Lin, L. Luo, Z. Zhao, Z. Li, Y. Zhang, Z. Yang, and J. Wang. 2017. An attention-based effective neural model for drug-drug interactions...