Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods

Hans Moen; Henry Suhonen; Sanna Salanterä; Tapio Salakoski; Laura Maria Peltonen

doi:10.3233/SHTI220606

Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods

Hans Moen^*, Henry Suhonen, Sanna Salanterä, Tapio Salakoski, Laura Maria Peltonen

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

3 Citations (Scopus)

80 Downloads (Pure)

Abstract

In health sciences, high-quality text embeddings may augment qualitative data analysis of large amounts of text by enabling, e.g., searching and clustering of health information. This study aimed to evaluate three different sentence-level embedding methods in clustering sentences in nursing narratives from individual patients' hospital care episodes. Two of these embeddings are generated from language models based on the BERT framework, and the third on the Sent2Vec method. These embedding methods were used to cluster sentences from 20 patient care episodes and the results were manually evaluated. Findings suggest that the best clusters were produced by the embeddings from a BERT model fine-tuned for the proxy task of predicting subject headings for nursing text.

Original language	English
Title of host publication	Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022
Editors	Brigitte Seroussi, Patrick Weber, Ferdinand Dhombres, Cyril Grouin, Jan-David Liebe, Jan-David Liebe, Jan-David Liebe, Sylvia Pelayo, Andrea Pinna, Bastien Rance, Bastien Rance, Lucia Sacchi, Adrien Ugon, Adrien Ugon, Arriel Benis, Parisis Gallos
Publisher	IOS Press
Pages	854-858
Number of pages	5
ISBN (Electronic)	978-1-64368-284-6
DOIs	https://doi.org/10.3233/SHTI220606
Publication status	Published - 25 May 2022
MoE publication type	A4 Conference publication
Event	Medical Informatics Europe Conference - Nice, France Duration: 27 May 2022 → 30 May 2022 Conference number: 32

Publication series

Name	Studies in Health Technology and Informatics
Publisher	IOS Press
Volume	294
ISSN (Print)	0926-9630
ISSN (Electronic)	1879-8365

Conference

Conference	Medical Informatics Europe Conference
Abbreviated title	MIE
Country/Territory	France
City	Nice
Period	27/05/2022 → 30/05/2022

Keywords

electronic health records
natural language processing
nursing documentation
sentence embeddings
Text clustering

Access to Document

10.3233/SHTI220606

Clustering Nursing SentencesFinal published version, 155 KBLicence: CC BY-NC

Cite this

Moen, H., Suhonen, H., Salanterä, S., Salakoski, T., & Peltonen, L. M. (2022). Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods. In B. Seroussi, P. Weber, F. Dhombres, C. Grouin, J.-D. Liebe, J.-D. Liebe, J.-D. Liebe, S. Pelayo, A. Pinna, B. Rance, B. Rance, L. Sacchi, A. Ugon, A. Ugon, A. Benis, & P. Gallos (Eds.), Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022 (pp. 854-858). (Studies in Health Technology and Informatics; Vol. 294). IOS Press. https://doi.org/10.3233/SHTI220606

Moen, Hans ; Suhonen, Henry ; Salanterä, Sanna et al. / Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods. Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022. editor / Brigitte Seroussi ; Patrick Weber ; Ferdinand Dhombres ; Cyril Grouin ; Jan-David Liebe ; Jan-David Liebe ; Jan-David Liebe ; Sylvia Pelayo ; Andrea Pinna ; Bastien Rance ; Bastien Rance ; Lucia Sacchi ; Adrien Ugon ; Adrien Ugon ; Arriel Benis ; Parisis Gallos. IOS Press, 2022. pp. 854-858 (Studies in Health Technology and Informatics).

@inproceedings{0144d4e87716434db9bbcfd6099e8719,

title = "Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods",

abstract = "In health sciences, high-quality text embeddings may augment qualitative data analysis of large amounts of text by enabling, e.g., searching and clustering of health information. This study aimed to evaluate three different sentence-level embedding methods in clustering sentences in nursing narratives from individual patients' hospital care episodes. Two of these embeddings are generated from language models based on the BERT framework, and the third on the Sent2Vec method. These embedding methods were used to cluster sentences from 20 patient care episodes and the results were manually evaluated. Findings suggest that the best clusters were produced by the embeddings from a BERT model fine-tuned for the proxy task of predicting subject headings for nursing text.",

keywords = "electronic health records, natural language processing, nursing documentation, sentence embeddings, Text clustering",

author = "Hans Moen and Henry Suhonen and Sanna Salanter{\"a} and Tapio Salakoski and Peltonen, {Laura Maria}",

note = "Funding Information: Acknowledgements: This work was supported by the Academy of Finland (grants 315376, 336033, 315896), Business Finland (grant 884/31/2018), and EU H2020 (grant 101016775). We thank Jari Bj{\"o}rne for helping with fine-tuning the BERT model. Publisher Copyright: {\textcopyright} 2022 European Federation for Medical Informatics (EFMI) and IOS Press. | openaire: EC/H2020/101016775/EU//INTERVENE ; Medical Informatics Europe Conference, MIE ; Conference date: 27-05-2022 Through 30-05-2022",

year = "2022",

month = may,

day = "25",

doi = "10.3233/SHTI220606",

language = "English",

series = "Studies in Health Technology and Informatics",

publisher = "IOS Press",

pages = "854--858",

editor = "Brigitte Seroussi and Patrick Weber and Ferdinand Dhombres and Cyril Grouin and Jan-David Liebe and Jan-David Liebe and Jan-David Liebe and Sylvia Pelayo and Andrea Pinna and Bastien Rance and Bastien Rance and Lucia Sacchi and Adrien Ugon and Adrien Ugon and Arriel Benis and Parisis Gallos",

booktitle = "Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022",

address = "Netherlands",

}

Moen, H, Suhonen, H, Salanterä, S, Salakoski, T & Peltonen, LM 2022, Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods. in B Seroussi, P Weber, F Dhombres, C Grouin, J-D Liebe, J-D Liebe, J-D Liebe, S Pelayo, A Pinna, B Rance, B Rance, L Sacchi, A Ugon, A Ugon, A Benis & P Gallos (eds), Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022. Studies in Health Technology and Informatics, vol. 294, IOS Press, pp. 854-858, Medical Informatics Europe Conference, Nice, France, 27/05/2022. https://doi.org/10.3233/SHTI220606

Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods. / Moen, Hans; Suhonen, Henry; Salanterä, Sanna et al.
Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022. ed. / Brigitte Seroussi; Patrick Weber; Ferdinand Dhombres; Cyril Grouin; Jan-David Liebe; Jan-David Liebe; Jan-David Liebe; Sylvia Pelayo; Andrea Pinna; Bastien Rance; Bastien Rance; Lucia Sacchi; Adrien Ugon; Adrien Ugon; Arriel Benis; Parisis Gallos. IOS Press, 2022. p. 854-858 (Studies in Health Technology and Informatics; Vol. 294).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods

AU - Moen, Hans

AU - Suhonen, Henry

AU - Salanterä, Sanna

AU - Salakoski, Tapio

AU - Peltonen, Laura Maria

N1 - Conference code: 32

PY - 2022/5/25

Y1 - 2022/5/25

N2 - In health sciences, high-quality text embeddings may augment qualitative data analysis of large amounts of text by enabling, e.g., searching and clustering of health information. This study aimed to evaluate three different sentence-level embedding methods in clustering sentences in nursing narratives from individual patients' hospital care episodes. Two of these embeddings are generated from language models based on the BERT framework, and the third on the Sent2Vec method. These embedding methods were used to cluster sentences from 20 patient care episodes and the results were manually evaluated. Findings suggest that the best clusters were produced by the embeddings from a BERT model fine-tuned for the proxy task of predicting subject headings for nursing text.

AB - In health sciences, high-quality text embeddings may augment qualitative data analysis of large amounts of text by enabling, e.g., searching and clustering of health information. This study aimed to evaluate three different sentence-level embedding methods in clustering sentences in nursing narratives from individual patients' hospital care episodes. Two of these embeddings are generated from language models based on the BERT framework, and the third on the Sent2Vec method. These embedding methods were used to cluster sentences from 20 patient care episodes and the results were manually evaluated. Findings suggest that the best clusters were produced by the embeddings from a BERT model fine-tuned for the proxy task of predicting subject headings for nursing text.

KW - electronic health records

KW - natural language processing

KW - nursing documentation

KW - sentence embeddings

KW - Text clustering

UR - http://www.scopus.com/inward/record.url?scp=85131105789&partnerID=8YFLogxK

U2 - 10.3233/SHTI220606

DO - 10.3233/SHTI220606

M3 - Conference article in proceedings

C2 - 35612225

AN - SCOPUS:85131105789

T3 - Studies in Health Technology and Informatics

SP - 854

EP - 858

BT - Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022

A2 - Seroussi, Brigitte

A2 - Weber, Patrick

A2 - Dhombres, Ferdinand

A2 - Grouin, Cyril

A2 - Liebe, Jan-David

A2 - Pelayo, Sylvia

A2 - Pinna, Andrea

A2 - Rance, Bastien

A2 - Sacchi, Lucia

A2 - Ugon, Adrien

A2 - Benis, Arriel

A2 - Gallos, Parisis

PB - IOS Press

T2 - Medical Informatics Europe Conference

Y2 - 27 May 2022 through 30 May 2022

ER -

Moen H, Suhonen H, Salanterä S, Salakoski T, Peltonen LM. Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods. In Seroussi B, Weber P, Dhombres F, Grouin C, Liebe JD, Liebe JD, Liebe JD, Pelayo S, Pinna A, Rance B, Rance B, Sacchi L, Ugon A, Ugon A, Benis A, Gallos P, editors, Challenges of Trustable AI and Added-Value on Health - Proceedings of MIE 2022. IOS Press. 2022. p. 854-858. (Studies in Health Technology and Informatics). doi: 10.3233/SHTI220606

Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

INTERVENE: International consortium for integrative genomics prediction

eMOM: CleverHealth Network: eMOM GDM -Project

Cite this

Clustering Nursing Sentences-Comparing Three Sentence Embedding Methods

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Projects

INTERVENE: International consortium for integrative genomics prediction

eMOM: CleverHealth Network: eMOM GDM -Project

Cite this