Patient Event Sequences for Predicting Hospitalization Length of Stay

Emil Riis Hansen; Thomas Dyhre Nielsen; Thomas Mulvad Larsen; Mads Nibe Strausholm; Tomer Sagi; Katja Hose

doi:10.1007/978-3-031-34344-5_7

Patient Event Sequences for Predicting Hospitalization Length of Stay

Emil Riis Hansen, Thomas Dyhre Nielsen, Thomas Mulvad Larsen, Mads Nibe Strausholm, Tomer Sagi, Katja Hose

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

Abstract

Predicting patients' hospital length of stay (LOS) is essential for improving resource allocation and supporting decision-making in healthcare organizations. This paper proposes a novel transformer-based model, termed Medic-BERT (M-BERT), for predicting LOS by modeling patient information as sequences of events. We performed empirical experiments on a cohort of $48k$ emergency care patients from a large Danish hospital. Experimental results show that M-BERT can achieve high accuracy on a variety of LOS problems and outperforms traditional non-sequence-based machine learning approaches.

Original language	English
Title of host publication	Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings
Editors	Jose M. Juarez, Mar Marcos, Gregor Stiglic, Allan Tucker
Number of pages	6
Volume	13897
Publisher	Springer, Cham
Publication date	7 Jun 2023
Pages	51-56
ISBN (Print)	9783031343438
DOIs	https://doi.org/10.1007/978-3-031-34344-5_7
Publication status	Published - 7 Jun 2023
Event	International Conference on Artificial Intelligence in Medicine - Bernardin Congress Centre, Portoroz, Slovenia Duration: 12 Jun 2023 → 15 Jun 2023 https://aime23.aimedicine.info/

Conference

Conference	International Conference on Artificial Intelligence in Medicine
Location	Bernardin Congress Centre
Country/Territory	Slovenia
City	Portoroz
Period	12/06/2023 → 15/06/2023
Internet address	https://aime23.aimedicine.info/

Keywords

length of stay prediction
sequence models
transformers

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-031-34344-5_7

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Hansen, E. R., Nielsen, T. D., Larsen, T. M., Strausholm, M. N., Sagi, T., & Hose, K. (2023). Patient Event Sequences for Predicting Hospitalization Length of Stay. In J. M. Juarez, M. Marcos, G. Stiglic, & A. Tucker (Eds.), Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings (Vol. 13897, pp. 51-56). Springer, Cham. https://doi.org/10.1007/978-3-031-34344-5_7

Hansen, Emil Riis ; Nielsen, Thomas Dyhre ; Larsen, Thomas Mulvad et al. / Patient Event Sequences for Predicting Hospitalization Length of Stay. Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings. editor / Jose M. Juarez ; Mar Marcos ; Gregor Stiglic ; Allan Tucker. Vol. 13897 Springer, Cham, 2023. pp. 51-56

@inproceedings{d864e8cf96324a729bf0facb0bab18fa,

title = "Patient Event Sequences for Predicting Hospitalization Length of Stay",

abstract = "Predicting patients' hospital length of stay (LOS) is essential for improving resource allocation and supporting decision-making in healthcare organizations. This paper proposes a novel transformer-based model, termed Medic-BERT (M-BERT), for predicting LOS by modeling patient information as sequences of events. We performed empirical experiments on a cohort of $48k$ emergency care patients from a large Danish hospital. Experimental results show that M-BERT can achieve high accuracy on a variety of LOS problems and outperforms traditional non-sequence-based machine learning approaches. ",

keywords = "length of stay prediction, sequence models, transformers",

author = "Hansen, {Emil Riis} and Nielsen, {Thomas Dyhre} and Larsen, {Thomas Mulvad} and Strausholm, {Mads Nibe} and Tomer Sagi and Katja Hose",

year = "2023",

month = jun,

day = "7",

doi = "10.1007/978-3-031-34344-5_7",

language = "English",

isbn = "9783031343438",

volume = "13897",

pages = "51--56",

editor = "Juarez, {Jose M.} and Mar Marcos and Gregor Stiglic and Allan Tucker",

booktitle = "Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings",

publisher = "Springer, Cham",

note = "International Conference on Artificial Intelligence in Medicine, AIME ; Conference date: 12-06-2023 Through 15-06-2023",

url = "https://aime23.aimedicine.info/",

}

Hansen, ER, Nielsen, TD, Larsen, TM, Strausholm, MN, Sagi, T & Hose, K 2023, Patient Event Sequences for Predicting Hospitalization Length of Stay. in JM Juarez, M Marcos, G Stiglic & A Tucker (eds), Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings. vol. 13897, Springer, Cham, pp. 51-56, International Conference on Artificial Intelligence in Medicine, Portoroz, Slovenia, 12/06/2023. https://doi.org/10.1007/978-3-031-34344-5_7

Patient Event Sequences for Predicting Hospitalization Length of Stay. / Hansen, Emil Riis; Nielsen, Thomas Dyhre; Larsen, Thomas Mulvad et al.
Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings. ed. / Jose M. Juarez; Mar Marcos; Gregor Stiglic; Allan Tucker. Vol. 13897 Springer, Cham, 2023. p. 51-56.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Patient Event Sequences for Predicting Hospitalization Length of Stay

AU - Hansen, Emil Riis

AU - Nielsen, Thomas Dyhre

AU - Larsen, Thomas Mulvad

AU - Strausholm, Mads Nibe

AU - Sagi, Tomer

AU - Hose, Katja

PY - 2023/6/7

Y1 - 2023/6/7

N2 - Predicting patients' hospital length of stay (LOS) is essential for improving resource allocation and supporting decision-making in healthcare organizations. This paper proposes a novel transformer-based model, termed Medic-BERT (M-BERT), for predicting LOS by modeling patient information as sequences of events. We performed empirical experiments on a cohort of $48k$ emergency care patients from a large Danish hospital. Experimental results show that M-BERT can achieve high accuracy on a variety of LOS problems and outperforms traditional non-sequence-based machine learning approaches.

AB - Predicting patients' hospital length of stay (LOS) is essential for improving resource allocation and supporting decision-making in healthcare organizations. This paper proposes a novel transformer-based model, termed Medic-BERT (M-BERT), for predicting LOS by modeling patient information as sequences of events. We performed empirical experiments on a cohort of $48k$ emergency care patients from a large Danish hospital. Experimental results show that M-BERT can achieve high accuracy on a variety of LOS problems and outperforms traditional non-sequence-based machine learning approaches.

KW - length of stay prediction

KW - sequence models

KW - transformers

UR - http://www.scopus.com/inward/record.url?scp=85151086836&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-34344-5_7

DO - 10.1007/978-3-031-34344-5_7

M3 - Article in proceeding

SN - 9783031343438

VL - 13897

SP - 51

EP - 56

BT - Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings

A2 - Juarez, Jose M.

A2 - Marcos, Mar

A2 - Stiglic, Gregor

A2 - Tucker, Allan

PB - Springer, Cham

T2 - International Conference on Artificial Intelligence in Medicine

Y2 - 12 June 2023 through 15 June 2023

ER -

Hansen ER, Nielsen TD, Larsen TM, Strausholm MN, Sagi T , Hose K. Patient Event Sequences for Predicting Hospitalization Length of Stay. In Juarez JM, Marcos M, Stiglic G, Tucker A, editors, Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Proceedings. Vol. 13897. Springer, Cham. 2023. p. 51-56 doi: 10.1007/978-3-031-34344-5_7

Patient Event Sequences for Predicting Hospitalization Length of Stay

Abstract

Conference

Keywords

UN SDGs

Access to Document

AUB Link

Other files and links

Fingerprint

Poul Due Jensen Professorate in Big Data and Artificial Intelligence

Cite this

Patient Event Sequences for Predicting Hospitalization Length of Stay

Abstract

Conference

Keywords

UN SDGs

Access to Document

AUB Link

Other files and links

Fingerprint

Projects

Poul Due Jensen Professorate in Big Data and Artificial Intelligence

Cite this