Cargando…
Using word embeddings to improve the privacy of clinical notes
OBJECTIVE: In this work, we introduce a privacy technique for anonymizing clinical notes that guarantees all private health information is secured (including sensitive data, such as family history, that are not adequately covered by current techniques). MATERIALS AND METHODS: We employ a new “random...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Oxford University Press
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7309261/ https://www.ncbi.nlm.nih.gov/pubmed/32388549 http://dx.doi.org/10.1093/jamia/ocaa038 |
_version_ | 1783549177324634112 |
---|---|
author | Abdalla, Mohamed Abdalla, Moustafa Rudzicz, Frank Hirst, Graeme |
author_facet | Abdalla, Mohamed Abdalla, Moustafa Rudzicz, Frank Hirst, Graeme |
author_sort | Abdalla, Mohamed |
collection | PubMed |
description | OBJECTIVE: In this work, we introduce a privacy technique for anonymizing clinical notes that guarantees all private health information is secured (including sensitive data, such as family history, that are not adequately covered by current techniques). MATERIALS AND METHODS: We employ a new “random replacement” paradigm (replacing each token in clinical notes with neighboring word vectors from the embedding space) to achieve 100% recall on the removal of sensitive information, unachievable with current “search-and-secure” paradigms. We demonstrate the utility of this paradigm on multiple corpora in a diverse set of classification tasks. RESULTS: We empirically evaluate the effect of our anonymization technique both on upstream and downstream natural language processing tasks to show that our perturbations, while increasing security (ie, achieving 100% recall on any dataset), do not greatly impact the results of end-to-end machine learning approaches. DISCUSSION: As long as current approaches utilize precision and recall to evaluate deidentification algorithms, there will remain a risk of overlooking sensitive information. Inspired by differential privacy, we sought to make it statistically infeasible to recreate the original data, although at the cost of readability. We hope that the work will serve as a catalyst to further research into alternative deidentification methods that can address current weaknesses. CONCLUSION: Our proposed technique can secure clinical texts at a low cost and extremely high recall with a readability trade-off while remaining useful for natural language processing classification tasks. We hope that our work can be used by risk-averse data holders to release clinical texts to researchers. |
format | Online Article Text |
id | pubmed-7309261 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | Oxford University Press |
record_format | MEDLINE/PubMed |
spelling | pubmed-73092612020-06-29 Using word embeddings to improve the privacy of clinical notes Abdalla, Mohamed Abdalla, Moustafa Rudzicz, Frank Hirst, Graeme J Am Med Inform Assoc Research and Applications OBJECTIVE: In this work, we introduce a privacy technique for anonymizing clinical notes that guarantees all private health information is secured (including sensitive data, such as family history, that are not adequately covered by current techniques). MATERIALS AND METHODS: We employ a new “random replacement” paradigm (replacing each token in clinical notes with neighboring word vectors from the embedding space) to achieve 100% recall on the removal of sensitive information, unachievable with current “search-and-secure” paradigms. We demonstrate the utility of this paradigm on multiple corpora in a diverse set of classification tasks. RESULTS: We empirically evaluate the effect of our anonymization technique both on upstream and downstream natural language processing tasks to show that our perturbations, while increasing security (ie, achieving 100% recall on any dataset), do not greatly impact the results of end-to-end machine learning approaches. DISCUSSION: As long as current approaches utilize precision and recall to evaluate deidentification algorithms, there will remain a risk of overlooking sensitive information. Inspired by differential privacy, we sought to make it statistically infeasible to recreate the original data, although at the cost of readability. We hope that the work will serve as a catalyst to further research into alternative deidentification methods that can address current weaknesses. CONCLUSION: Our proposed technique can secure clinical texts at a low cost and extremely high recall with a readability trade-off while remaining useful for natural language processing classification tasks. We hope that our work can be used by risk-averse data holders to release clinical texts to researchers. Oxford University Press 2020-05-10 /pmc/articles/PMC7309261/ /pubmed/32388549 http://dx.doi.org/10.1093/jamia/ocaa038 Text en © The Author(s) 2020. Published by Oxford University Press on behalf of the American Medical Informatics Association. http://creativecommons.org/licenses/by-nc/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com |
spellingShingle | Research and Applications Abdalla, Mohamed Abdalla, Moustafa Rudzicz, Frank Hirst, Graeme Using word embeddings to improve the privacy of clinical notes |
title | Using word embeddings to improve the privacy of clinical notes |
title_full | Using word embeddings to improve the privacy of clinical notes |
title_fullStr | Using word embeddings to improve the privacy of clinical notes |
title_full_unstemmed | Using word embeddings to improve the privacy of clinical notes |
title_short | Using word embeddings to improve the privacy of clinical notes |
title_sort | using word embeddings to improve the privacy of clinical notes |
topic | Research and Applications |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7309261/ https://www.ncbi.nlm.nih.gov/pubmed/32388549 http://dx.doi.org/10.1093/jamia/ocaa038 |
work_keys_str_mv | AT abdallamohamed usingwordembeddingstoimprovetheprivacyofclinicalnotes AT abdallamoustafa usingwordembeddingstoimprovetheprivacyofclinicalnotes AT rudziczfrank usingwordembeddingstoimprovetheprivacyofclinicalnotes AT hirstgraeme usingwordembeddingstoimprovetheprivacyofclinicalnotes |