Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation

Benjamin C. Knoll; Elizabeth A. Lindemann; Arian L. Albert; Genevieve B. Melton; Serguei V.S. Pakhomov

doi:10.3233/SHTI190211

Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation

Benjamin C. Knoll, Elizabeth A. Lindemann, Arian L. Albert, Genevieve B. Melton, Serguei V.S. Pakhomov

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Scopus citations

Abstract

Although a number of foundational natural language processing (NLP) tasks like text segmentation are considered a simple problem in the general English domain dominated by well-formed text, complexities of clinical documentation lead to poor performance of existing solutions designed for the general English domain. We present an alternative solution that relies on a convolutional neural network layer followed by a bidirectional long short-term memory layer (CNN-Bi-LSTM) for the task of sentence boundary disambiguation and describe an ensemble approach for domain adaptation using two training corpora. Implementations using the Keras neural-networks API are available at https://github.com/NLPIE/clinical-sentences.

Original language	English (US)
Title of host publication	MEDINFO 2019
Subtitle of host publication	Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics
Editors	Brigitte Seroussi, Lucila Ohno-Machado, Lucila Ohno-Machado, Brigitte Seroussi
Publisher	IOS Press
Pages	198-202
Number of pages	5
Volume	264
ISBN (Electronic)	9781643680026
DOIs	https://doi.org/10.3233/SHTI190211
State	Published - Aug 21 2019
Event	17th World Congress on Medical and Health Informatics, MEDINFO 2019 - Lyon, France Duration: Aug 25 2019 → Aug 30 2019

Publication series

Name	Studies in health technology and informatics
ISSN (Print)	0926-9630

Conference

Conference	17th World Congress on Medical and Health Informatics, MEDINFO 2019
Country/Territory	France
City	Lyon
Period	8/25/19 → 8/30/19

Bibliographical note

Funding Information:
We would like to acknowledge Michael Hietpas, one of the annotators for our data. This research was supported in part by NIH/NCATS UL1TR002494, NIH/NCATS U01TR002062, NIH/NIGMS R01GM102282, and AHRQ R01HS022085. The content is solely the responsibility of the authors and does not necessary represent the official views of the National Institutes of Health or the Agency for Healthcare Research and Quality.

Publisher Copyright:
© 2019 International Medical Informatics Association (IMIA) and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).

Keywords

Machine Learning
Natural Language Processing
Neural Networks (Computer)
Neural Networks, Computer
Language
Documentation

PubMed: MeSH publication types

Journal Article

Publisher link

10.3233/SHTI190211

Find It

Find It at U of M Libraries

Cite this

Knoll, B. C., Lindemann, E. A., Albert, A. L., Melton, G. B., & Pakhomov, S. V. S. (2019). Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation. In B. Seroussi, L. Ohno-Machado, L. Ohno-Machado, & B. Seroussi (Eds.), MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics (Vol. 264, pp. 198-202). (Studies in health technology and informatics). IOS Press. https://doi.org/10.3233/SHTI190211

Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation. / Knoll, Benjamin C.; Lindemann, Elizabeth A.; Albert, Arian L. et al.
MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. ed. / Brigitte Seroussi; Lucila Ohno-Machado; Lucila Ohno-Machado; Brigitte Seroussi. Vol. 264 IOS Press, 2019. p. 198-202 (Studies in health technology and informatics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Knoll, BC, Lindemann, EA, Albert, AL, Melton, GB & Pakhomov, SVS 2019, Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation. in B Seroussi, L Ohno-Machado, L Ohno-Machado & B Seroussi (eds), MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. vol. 264, Studies in health technology and informatics, IOS Press, pp. 198-202, 17th World Congress on Medical and Health Informatics, MEDINFO 2019, Lyon, France, 8/25/19. https://doi.org/10.3233/SHTI190211

Knoll BC, Lindemann EA, Albert AL, Melton GB , Pakhomov SVS. Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation. In Seroussi B, Ohno-Machado L, Ohno-Machado L, Seroussi B, editors, MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. Vol. 264. IOS Press. 2019. p. 198-202. (Studies in health technology and informatics). doi: 10.3233/SHTI190211

Knoll, Benjamin C. ; Lindemann, Elizabeth A. ; Albert, Arian L. et al. / Recurrent deep network models for clinical NLP tasks : Use case with sentence boundary disambiguation. MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. editor / Brigitte Seroussi ; Lucila Ohno-Machado ; Lucila Ohno-Machado ; Brigitte Seroussi. Vol. 264 IOS Press, 2019. pp. 198-202 (Studies in health technology and informatics).

@inproceedings{8da26c93d78a49c685e5231852d24e64,

title = "Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation",

abstract = "Although a number of foundational natural language processing (NLP) tasks like text segmentation are considered a simple problem in the general English domain dominated by well-formed text, complexities of clinical documentation lead to poor performance of existing solutions designed for the general English domain. We present an alternative solution that relies on a convolutional neural network layer followed by a bidirectional long short-term memory layer (CNN-Bi-LSTM) for the task of sentence boundary disambiguation and describe an ensemble approach for domain adaptation using two training corpora. Implementations using the Keras neural-networks API are available at https://github.com/NLPIE/clinical-sentences.",

keywords = "Machine Learning, Natural Language Processing, Neural Networks (Computer), Neural Networks, Computer, Language, Documentation",

author = "Knoll, {Benjamin C.} and Lindemann, {Elizabeth A.} and Albert, {Arian L.} and Melton, {Genevieve B.} and Pakhomov, {Serguei V.S.}",

note = "Funding Information: We would like to acknowledge Michael Hietpas, one of the annotators for our data. This research was supported in part by NIH/NCATS UL1TR002494, NIH/NCATS U01TR002062, NIH/NIGMS R01GM102282, and AHRQ R01HS022085. The content is solely the responsibility of the authors and does not necessary represent the official views of the National Institutes of Health or the Agency for Healthcare Research and Quality. Publisher Copyright: {\textcopyright} 2019 International Medical Informatics Association (IMIA) and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).; 17th World Congress on Medical and Health Informatics, MEDINFO 2019 ; Conference date: 25-08-2019 Through 30-08-2019",

year = "2019",

month = aug,

day = "21",

doi = "10.3233/SHTI190211",

language = "English (US)",

volume = "264",

series = "Studies in health technology and informatics",

publisher = "IOS Press",

pages = "198--202",

editor = "Brigitte Seroussi and Lucila Ohno-Machado and Lucila Ohno-Machado and Brigitte Seroussi",

booktitle = "MEDINFO 2019",

}

TY - GEN

T1 - Recurrent deep network models for clinical NLP tasks

T2 - 17th World Congress on Medical and Health Informatics, MEDINFO 2019

AU - Knoll, Benjamin C.

AU - Lindemann, Elizabeth A.

AU - Albert, Arian L.

AU - Melton, Genevieve B.

AU - Pakhomov, Serguei V.S.

N1 - Funding Information: We would like to acknowledge Michael Hietpas, one of the annotators for our data. This research was supported in part by NIH/NCATS UL1TR002494, NIH/NCATS U01TR002062, NIH/NIGMS R01GM102282, and AHRQ R01HS022085. The content is solely the responsibility of the authors and does not necessary represent the official views of the National Institutes of Health or the Agency for Healthcare Research and Quality. Publisher Copyright: © 2019 International Medical Informatics Association (IMIA) and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).

PY - 2019/8/21

Y1 - 2019/8/21

N2 - Although a number of foundational natural language processing (NLP) tasks like text segmentation are considered a simple problem in the general English domain dominated by well-formed text, complexities of clinical documentation lead to poor performance of existing solutions designed for the general English domain. We present an alternative solution that relies on a convolutional neural network layer followed by a bidirectional long short-term memory layer (CNN-Bi-LSTM) for the task of sentence boundary disambiguation and describe an ensemble approach for domain adaptation using two training corpora. Implementations using the Keras neural-networks API are available at https://github.com/NLPIE/clinical-sentences.

AB - Although a number of foundational natural language processing (NLP) tasks like text segmentation are considered a simple problem in the general English domain dominated by well-formed text, complexities of clinical documentation lead to poor performance of existing solutions designed for the general English domain. We present an alternative solution that relies on a convolutional neural network layer followed by a bidirectional long short-term memory layer (CNN-Bi-LSTM) for the task of sentence boundary disambiguation and describe an ensemble approach for domain adaptation using two training corpora. Implementations using the Keras neural-networks API are available at https://github.com/NLPIE/clinical-sentences.

KW - Machine Learning

KW - Natural Language Processing

KW - Neural Networks (Computer)

KW - Neural Networks, Computer

KW - Language

KW - Documentation

UR - http://www.scopus.com/inward/record.url?scp=85071481529&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071481529&partnerID=8YFLogxK

U2 - 10.3233/SHTI190211

DO - 10.3233/SHTI190211

M3 - Conference contribution

C2 - 31437913

AN - SCOPUS:85071481529

VL - 264

T3 - Studies in health technology and informatics

SP - 198

EP - 202

BT - MEDINFO 2019

A2 - Seroussi, Brigitte

A2 - Ohno-Machado, Lucila

A2 - Seroussi, Brigitte

PB - IOS Press

Y2 - 25 August 2019 through 30 August 2019

ER -

Recurrent deep network models for clinical NLP tasks: Use case with sentence boundary disambiguation

Abstract

Publication series

Conference

Bibliographical note

Keywords

PubMed: MeSH publication types

Publisher link

Find It

Other files and links

Fingerprint

Cite this