Sensitive Data Detection and Classification in Spanish Clinical Text: Experiments with BERT

Massive digital data processing provides a wide range of opportunities and benefits, but at the cost of endangering personal data privacy. Anonymisation consists in removing or replacing sensitive information from data, enabling its exploitation for different purposes while preserving the privacy of individuals. Over the years, a lot of automatic anonymisation systems have been proposed; however, depending on the type of data, the target language or the availability of training documents, the task remains challenging still. The emergence of novel deep-learning models during the last two years has brought large improvements to the state of the art in the field of Natural Language Processing. These advancements have been most noticeably led by BERT, a model proposed by Google in 2018, and the shared language models pre-trained on millions of documents. In this paper, we use a BERT-based sequence labelling model to conduct a series of anonymisation experiments on several clinical datasets in Spanish. We also compare BERT to other algorithms. The experiments show that a simple BERT-based model with general-domain pre-training obtains highly competitive results without any domain specific feature engineering.

[1]  Frank Hutter,et al.  Decoupled Weight Decay Regularization , 2017, ICLR.

[2]  Veselin Stoyanov,et al.  Unsupervised Cross-lingual Representation Learning at Scale , 2019, ACL.

[3]  Rema Padman,et al.  A Deep Learning Architecture for De-identification of Patient Notes: Implementation and Evaluation , 2018, ArXiv.

[4]  David Yarowsky,et al.  Techniques in Speech Acoustics , 1999, Computational Linguistics.

[5]  Amund Tveita,et al.  Anonymization of General Practioner Medical Records , 2004 .

[6]  Aitor Gonzalez-Agirre,et al.  Automatic De-identification of Medical Texts in Spanish: the MEDDOCAN Track, Corpus, Guidelines, Methods and Evaluation of Results , 2019, IberLEF@SEPLN.

[7]  Manex Serras,et al.  Vicomtech at MEDDOCAN: Medical Document Anonymization , 2019, IberLEF@SEPLN.

[8]  Yiming Yang,et al.  XLNet: Generalized Autoregressive Pretraining for Language Understanding , 2019, NeurIPS.

[9]  Dou Shen Text Summarization , 2009, Encyclopedia of Database Systems.

[10]  R'emi Louf,et al.  HuggingFace's Transformers: State-of-the-art Natural Language Processing , 2019, ArXiv.

[11]  Mirella Lapata,et al.  Text Summarization with Pretrained Encoders , 2019, EMNLP.

[12]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[13]  Franck Dernoncourt,et al.  De-identification of patient notes with recurrent neural networks , 2016, J. Am. Medical Informatics Assoc..

[14]  Kyunghyun Cho,et al.  Passage Re-ranking with BERT , 2019, ArXiv.

[15]  Mark Dredze,et al.  Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERT , 2019, EMNLP.

[16]  Sampo Pyysalo,et al.  brat: a Web-based Tool for NLP-Assisted Text Annotation , 2012, EACL.

[17]  Wanli Liu,et al.  Hadoken: a BERT-CRF Model for Medical Document Anonymization , 2019, IberLEF@SEPLN.

[18]  Omer Levy,et al.  RoBERTa: A Robustly Optimized BERT Pretraining Approach , 2019, ArXiv.

[19]  Peter Szolovits,et al.  Evaluating the state-of-the-art in automatic de-identification. , 2007, Journal of the American Medical Informatics Association : JAMIA.

[20]  Laura García Sardiña Automating the anonymisation of textual corpora , 2018 .

[21]  Heike Adel,et al.  NLNDE: The Neither-Language-Nor-Domain-Experts' Way of Spanish Medical Document De-Identification , 2020, IberLEF@SEPLN.

[22]  Abderrahim Beni Hssane,et al.  Big healthcare data: preserving security and privacy , 2018, Journal of Big Data.

[23]  Özlem Uzuner,et al.  Automated systems for the de-identification of longitudinal clinical narratives: Overview of 2014 i2b2/UTHealth shared task Track 1 , 2015, J. Biomed. Informatics.

[24]  German Rigau,et al.  IXA pipeline: Efficient and Ready to Use Multilingual NLP tools , 2014, LREC.

[25]  Andrew McCallum,et al.  Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[26]  Jorge Baptista,et al.  Automated anonymization of text documents , 2016, 2016 IEEE Congress on Evolutionary Computation (CEC).

[27]  Jorge Turmo Borras,et al.  Building a Spanish/Catalan health records corpus with very sparse protected information labelled , 2018 .

[28]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[29]  Andrew McCallum,et al.  Fast and Accurate Entity Recognition with Iterated Dilated Convolutions , 2017, EMNLP.

[30]  Montse Cuadros,et al.  NUBes: A Corpus of Negation and Uncertainty in Spanish Clinical Texts , 2020, LREC.

[31]  Ilya Sutskever,et al.  Language Models are Unsupervised Multitask Learners , 2019 .