Interpretable and continuous prediction of acute kidney injury in the intensive care

Iacopo Vagliano; Oleksandra Lvova; Martijn C. Schut

doi:https://doi.org/10.3233/SHTI210129

Interpretable and continuous prediction of acute kidney injury in the intensive care

Iacopo Vagliano, Oleksandra Lvova, Martijn C. Schut

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic › peer-review

5 Citations (Scopus)

Abstract

Acute kidney injury (AKI) is a common and potentially life-threatening condition, which often occurs in the intensive care unit. We propose a machine learning model based on recurrent neural networks to continuously predict AKI. We internally validated its predictive performance, both in terms of discrimination and calibration, and assessed its interpretability. Our model achieved good discrimination (AUC 0.80-0.94). Such a continuous model can support clinicians to promptly recognize and treat AKI patients and may improve their outcomes.

Original language	English
Title of host publication	Public Health and Informatics
Subtitle of host publication	Proceedings of MIE 2021
Publisher	IOS Press
Pages	103-107
Number of pages	5
ISBN (Electronic)	9781643681856
ISBN (Print)	9781643681849
DOIs	https://doi.org/10.3233/SHTI210129
Publication status	Published - 1 Jul 2021

Keywords

Acute kidney injury
Clinical prediction models
ICU
Machine learning

Access to Document

https://doi.org/10.3233/SHTI210129

Cite this

@inbook{af404009e2fb4ec6a4246dd98f3c441c,

title = "Interpretable and continuous prediction of acute kidney injury in the intensive care",

abstract = "Acute kidney injury (AKI) is a common and potentially life-threatening condition, which often occurs in the intensive care unit. We propose a machine learning model based on recurrent neural networks to continuously predict AKI. We internally validated its predictive performance, both in terms of discrimination and calibration, and assessed its interpretability. Our model achieved good discrimination (AUC 0.80-0.94). Such a continuous model can support clinicians to promptly recognize and treat AKI patients and may improve their outcomes.",

keywords = "Acute kidney injury, Clinical prediction models, ICU, Machine learning",

author = "Iacopo Vagliano and Oleksandra Lvova and Schut, {Martijn C.}",

note = "Publisher Copyright: {\textcopyright} 2021 European Federation for Medical Informatics (EFMI) and IOS Press. All rights reserved. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.",

year = "2021",

month = jul,

day = "1",

doi = "https://doi.org/10.3233/SHTI210129",

language = "English",

isbn = "9781643681849",

pages = "103--107",

booktitle = "Public Health and Informatics",

publisher = "IOS Press",

address = "Netherlands",

}

TY - CHAP

T1 - Interpretable and continuous prediction of acute kidney injury in the intensive care

AU - Vagliano, Iacopo

AU - Lvova, Oleksandra

AU - Schut, Martijn C.

PY - 2021/7/1

Y1 - 2021/7/1

N2 - Acute kidney injury (AKI) is a common and potentially life-threatening condition, which often occurs in the intensive care unit. We propose a machine learning model based on recurrent neural networks to continuously predict AKI. We internally validated its predictive performance, both in terms of discrimination and calibration, and assessed its interpretability. Our model achieved good discrimination (AUC 0.80-0.94). Such a continuous model can support clinicians to promptly recognize and treat AKI patients and may improve their outcomes.

AB - Acute kidney injury (AKI) is a common and potentially life-threatening condition, which often occurs in the intensive care unit. We propose a machine learning model based on recurrent neural networks to continuously predict AKI. We internally validated its predictive performance, both in terms of discrimination and calibration, and assessed its interpretability. Our model achieved good discrimination (AUC 0.80-0.94). Such a continuous model can support clinicians to promptly recognize and treat AKI patients and may improve their outcomes.

KW - Acute kidney injury

KW - Clinical prediction models

KW - ICU

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85107238178&partnerID=8YFLogxK

U2 - https://doi.org/10.3233/SHTI210129

DO - https://doi.org/10.3233/SHTI210129

M3 - Chapter

C2 - 34042714

SN - 9781643681849

SP - 103

EP - 107

BT - Public Health and Informatics

PB - IOS Press

ER -

Interpretable and continuous prediction of acute kidney injury in the intensive care

Abstract

Keywords

Access to Document

Other files and links

Cite this