Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery

Linda Lapp; Marc Roper; Kimberley Kavanagh; Stefan Schraag

doi:10.1109/cbms55023.2022.00048

Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery

Linda Lapp, Marc Roper, Kimberley Kavanagh, Stefan Schraag

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

2 Citations (Scopus)

28 Downloads (Pure)

Abstract

Delirium, affecting up to 52% of cardiac surgery patients, can have serious long-term effects on patients by damaging cognitive ability and causing subsequent functional decline. This study reports on the development and evaluation of predictive models aimed at identifying the likely onset of delirium on an hourly basis in intensive care unit following cardiac surgery. Most models achieved a mean AUC > 0.900 across all lead times. A support vector machine achieved the highest performance across all lead times of AUC = 0.941 and Sensitivity = 0.907, and BARTm, where missing values were replaced with missForest imputation, achieved the highest Specificity of 0.892. Being able to predict delirium hours in advance gives clinicians the ability to intervene and optimize treatments for patients who are at risk and avert potentially serious and life-threatening consequences.

Original language	English
Title of host publication	2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS)
Editors	Linlin Shen, Alejandro Rodriguez Gonzalez, KC Santosh, Zhihui Lai, Rosa Sicilia, Joao Rafael Almeida, Bridget Kane
Place of Publication	Piscataway, N.J.
Publisher	IEEE
Pages	234-239
Number of pages	6
ISBN (Electronic)	9781665467704
DOIs	https://doi.org/10.1109/cbms55023.2022.00048
Publication status	Published - 31 Aug 2022

Keywords

medical computing
patient monitoring
support vector machines
intensive care unit

Access to Document

10.1109/cbms55023.2022.00048

Lapp-etal-IEEE-CBMS-2022-Predicting-the-onset-of-delirium-on-hourlyAccepted author manuscript, 965 KBLicence: Strath_1

Code for: "Predicting the Onset of Delirium on Hourly Basis in an Intensive Care Unit Following Cardiac Surgery"
Lapp, L. (Creator) & Roper, M. (Supervisor), University of Strathclyde, 14 Jun 2022
DOI: 10.15129/1ab360f7-0779-4cf3-8a9a-dae621892a51, https://2022.cbms-conference.org/
Dataset

Cite this

Lapp, L., Roper, M., Kavanagh, K., & Schraag, S. (2022). Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery. In L. Shen, A. R. Gonzalez, KC. Santosh, Z. Lai, R. Sicilia, J. R. Almeida, & B. Kane (Eds.), 2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS) (pp. 234-239). IEEE. https://doi.org/10.1109/cbms55023.2022.00048

Lapp, Linda ; Roper, Marc ; Kavanagh, Kimberley et al. / Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery. 2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS). editor / Linlin Shen ; Alejandro Rodriguez Gonzalez ; KC Santosh ; Zhihui Lai ; Rosa Sicilia ; Joao Rafael Almeida ; Bridget Kane. Piscataway, N.J. : IEEE, 2022. pp. 234-239

@inproceedings{36a1236f34264c4ebd1f67d7509ccfcd,

title = "Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery",

abstract = "Delirium, affecting up to 52% of cardiac surgery patients, can have serious long-term effects on patients by damaging cognitive ability and causing subsequent functional decline. This study reports on the development and evaluation of predictive models aimed at identifying the likely onset of delirium on an hourly basis in intensive care unit following cardiac surgery. Most models achieved a mean AUC > 0.900 across all lead times. A support vector machine achieved the highest performance across all lead times of AUC = 0.941 and Sensitivity = 0.907, and BARTm, where missing values were replaced with missForest imputation, achieved the highest Specificity of 0.892. Being able to predict delirium hours in advance gives clinicians the ability to intervene and optimize treatments for patients who are at risk and avert potentially serious and life-threatening consequences.",

keywords = "medical computing, patient monitoring, support vector machines, intensive care unit",

author = "Linda Lapp and Marc Roper and Kimberley Kavanagh and Stefan Schraag",

note = "{\textcopyright} 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",

year = "2022",

month = aug,

day = "31",

doi = "10.1109/cbms55023.2022.00048",

language = "English",

pages = "234--239",

editor = "Linlin Shen and Gonzalez, {Alejandro Rodriguez} and KC Santosh and Zhihui Lai and Rosa Sicilia and Almeida, {Joao Rafael} and Bridget Kane",

booktitle = "2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS)",

publisher = "IEEE",

}

Lapp, L , Roper, M , Kavanagh, K & Schraag, S 2022, Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery. in L Shen, AR Gonzalez, KC Santosh, Z Lai, R Sicilia, JR Almeida & B Kane (eds), 2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS). IEEE, Piscataway, N.J., pp. 234-239. https://doi.org/10.1109/cbms55023.2022.00048

Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery. / Lapp, Linda ; Roper, Marc ; Kavanagh, Kimberley et al.
2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS). ed. / Linlin Shen; Alejandro Rodriguez Gonzalez; KC Santosh; Zhihui Lai; Rosa Sicilia; Joao Rafael Almeida; Bridget Kane. Piscataway, N.J.: IEEE, 2022. p. 234-239.

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

TY - GEN

T1 - Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery

AU - Lapp, Linda

AU - Roper, Marc

AU - Kavanagh, Kimberley

AU - Schraag, Stefan

N1 - © 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2022/8/31

Y1 - 2022/8/31

N2 - Delirium, affecting up to 52% of cardiac surgery patients, can have serious long-term effects on patients by damaging cognitive ability and causing subsequent functional decline. This study reports on the development and evaluation of predictive models aimed at identifying the likely onset of delirium on an hourly basis in intensive care unit following cardiac surgery. Most models achieved a mean AUC > 0.900 across all lead times. A support vector machine achieved the highest performance across all lead times of AUC = 0.941 and Sensitivity = 0.907, and BARTm, where missing values were replaced with missForest imputation, achieved the highest Specificity of 0.892. Being able to predict delirium hours in advance gives clinicians the ability to intervene and optimize treatments for patients who are at risk and avert potentially serious and life-threatening consequences.

AB - Delirium, affecting up to 52% of cardiac surgery patients, can have serious long-term effects on patients by damaging cognitive ability and causing subsequent functional decline. This study reports on the development and evaluation of predictive models aimed at identifying the likely onset of delirium on an hourly basis in intensive care unit following cardiac surgery. Most models achieved a mean AUC > 0.900 across all lead times. A support vector machine achieved the highest performance across all lead times of AUC = 0.941 and Sensitivity = 0.907, and BARTm, where missing values were replaced with missForest imputation, achieved the highest Specificity of 0.892. Being able to predict delirium hours in advance gives clinicians the ability to intervene and optimize treatments for patients who are at risk and avert potentially serious and life-threatening consequences.

KW - medical computing

KW - patient monitoring

KW - support vector machines

KW - intensive care unit

U2 - 10.1109/cbms55023.2022.00048

DO - 10.1109/cbms55023.2022.00048

M3 - Conference contribution book

SP - 234

EP - 239

BT - 2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS)

A2 - Shen, Linlin

A2 - Gonzalez, Alejandro Rodriguez

A2 - Santosh, KC

A2 - Lai, Zhihui

A2 - Sicilia, Rosa

A2 - Almeida, Joao Rafael

A2 - Kane, Bridget

PB - IEEE

CY - Piscataway, N.J.

ER -

Lapp L , Roper M , Kavanagh K, Schraag S. Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery. In Shen L, Gonzalez AR, Santosh KC, Lai Z, Sicilia R, Almeida JR, Kane B, editors, 2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS). Piscataway, N.J.: IEEE. 2022. p. 234-239 doi: 10.1109/cbms55023.2022.00048

Predicting the onset of delirium on hourly basis in an intensive care unit following cardiac surgery

Abstract

Keywords

Access to Document

Fingerprint

Datasets

Code for: "Predicting the Onset of Delirium on Hourly Basis in an Intensive Care Unit Following Cardiac Surgery"

Cite this