Journal article · 2022 · Open access

Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model

Seo Hyun Oh, Min Kang, Youngho Lee

Healthcare Informatics Research, 28(1), 16–24 · Published 31 January 2022 · DOI 10.4258/hir.2022.28.1.16

Open PDFView at the publisherCite this paper

In brief

“Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model” is a 2022 journal article by Seo Hyun Oh and 2 co-authors, published in Healthcare Informatics Research. OpenAlex counts 20 citations as of 19 November 2026, which puts it in the top 5% of works published in 2022 by citations. It is open access under a CC BY-NC licence: the full text is free to read at Healthcare Informatics Research.

Citations
20
Most cited in
2024 (6)
Field-weighted impact
1.06
References
17
Authors
3

Abstract

OBJECTIVE: De-identifying protected health information (PHI) in medical documents is important, and a prerequisite to deidentification is the identification of PHI entity names in clinical documents. This study aimed to compare the performance of three pre-training models that have recently attracted significant attention and to determine which model is more suitable for PHI recognition. METHODS: We compared the PHI recognition performance of deep learning models using the i2b2 2014 dataset. We used the three pre-training models-namely, bidirectional encoder representations from transformers (BERT), robustly optimized BERT pre-training approach (RoBERTa), and XLNet (model built based on Transformer-XL)-to detect PHI. After the dataset was tokenized, it was processed using an inside-outside-beginning tagging scheme and WordPiecetokenized to place it into these models. Further, the PHI recognition performance was investigated using BERT, RoBERTa, and XLNet. RESULTS: Comparing the PHI recognition performance of the three models, it was confirmed that XLNet had a superior F1-score of 96.29%. In addition, when checking PHI entity performance evaluation, RoBERTa and XLNet showed a 30% improvement in performance compared to BERT. CONCLUSIONS: Among the pre-training models used in this study, XLNet exhibited superior performance because word embedding was well constructed using the two-stream self-attention method. In addition, compared to BERT, RoBERTa and XLNet showed superior performance, indicating that they were more effective in grasping the context.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Oh, S. H., Kang, M., & Lee, Y. (2022). Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model. Healthcare Informatics Research, 28(1), 16–24. https://doi.org/10.4258/hir.2022.28.1.16

MLA 9

Oh, Seo Hyun, et al. “Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model.” Healthcare Informatics Research, vol. 28, no. 1, 2022, pp. 16–24. https://doi.org/10.4258/hir.2022.28.1.16.

Chicago

Oh, Seo Hyun, Min Kang, and Youngho Lee. 2022. “Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model.” Healthcare Informatics Research 28 (1): 16–24. https://doi.org/10.4258/hir.2022.28.1.16.

Harvard

Oh, S. H., Kang, M. and Lee, Y. (2022) ‘Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model’, Healthcare Informatics Research, 28(1), pp. 16–24. doi:10.4258/hir.2022.28.1.16.

IEEE

S. H. Oh, M. Kang, and Y. Lee, “Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model,” Healthcare Informatics Research, vol. 28, no. 1, pp. 16–24, 2022, doi: 10.4258/hir.2022.28.1.16.

BibTeX

@article{oh2022protected,
  title     = {{Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model}},
  author    = {Oh, Seo Hyun and Kang, Min and Lee, Youngho},
  journal   = {Healthcare Informatics Research},
  year      = {2022},
  volume    = {28},
  number    = {1},
  pages     = {16--24},
  doi       = {10.4258/hir.2022.28.1.16},
  url       = {https://doi.org/10.4258/hir.2022.28.1.16}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
202233
202333
202466
202555
202633

Its field-weighted citation impact is 1.06: it has been cited 1.06 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling

  • de-identification
  • clinical text
  • BERT
  • RoBERTa
  • XLNet
  • Transformer-XL

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Oh, S. H., Kang, M., & Lee, Y. (2022). Protected Health Information Recognition by Fine-Tuning a Pre-training Transformer Model. Healthcare Informatics Research, 28(1), 16–24. https://doi.org/10.4258/hir.2022.28.1.16

How many times has this paper been cited?

20 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Healthcare Informatics Research: sign in and open the PDF from this page.

Who wrote this paper?

Seo Hyun Oh, Min Kang, Youngho Lee.

What is the DOI of this paper?

10.4258/hir.2022.28.1.16, permanently at https://doi.org/10.4258/hir.2022.28.1.16. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.