Towards personalized intensive care decision support using a Bayesian network

A multicenter glycemic control study

Asma Abu-Samah, Normy Norfiza Abdul Razak, Fatanah Mohamad Suhaimi, Ummu Kulthum Jamaludin, James Geoffrey Chase

Research output: Contribution to journalArticle

Abstract

Personalized treatment in glycemic control (GC) is a visibly promising research area that requires improved mechanisms providing patient-specific procedures to enable complicated decision support. Available per-patient data must be more than written records, and be fully integrated in this personalization process. This article presents a process for relating the intensive care unit patients' demographic and admission data to their GC performance. With this objective, a probabilistic Bayesian network was chosen to provide more personalized decisions. As a case study, average daily blood glucose measurements were chosen as the interest target node in order to weigh GC that provides a reduced nursing workload. To test the idea, data from 482 patients, with nine variables from four Malaysian intensive care units with different controls were exploited. The identified steps crucial in building a dependable model are variable selection, continuous state discretization, and unsupervised structure learning. Using a multi-target node evaluation, a network with 80% mean overall classification precision was obtained with a normalized equal distance discretization algorithm and a maximum weight spanning tree technique. Meanwhile, the interest target node scored 90.39% precision. The results from this study, which are complemented with an evaluation of missing data, are proposed as a benchmark for using Bayesian networks in this type of application.

Original languageEnglish
Pages (from-to)202-209
Number of pages8
JournalIEIE Transactions on Smart Processing and Computing
Volume8
Issue number3
DOIs
Publication statusPublished - 01 Jan 2019

Fingerprint

Bayesian networks
Intensive care units
Nursing
Glucose
Blood

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Abu-Samah, Asma ; Abdul Razak, Normy Norfiza ; Suhaimi, Fatanah Mohamad ; Jamaludin, Ummu Kulthum ; Chase, James Geoffrey. / Towards personalized intensive care decision support using a Bayesian network : A multicenter glycemic control study. In: IEIE Transactions on Smart Processing and Computing. 2019 ; Vol. 8, No. 3. pp. 202-209.
@article{153ef7185fa64b6c9690b60ef0e089ac,
title = "Towards personalized intensive care decision support using a Bayesian network: A multicenter glycemic control study",
abstract = "Personalized treatment in glycemic control (GC) is a visibly promising research area that requires improved mechanisms providing patient-specific procedures to enable complicated decision support. Available per-patient data must be more than written records, and be fully integrated in this personalization process. This article presents a process for relating the intensive care unit patients' demographic and admission data to their GC performance. With this objective, a probabilistic Bayesian network was chosen to provide more personalized decisions. As a case study, average daily blood glucose measurements were chosen as the interest target node in order to weigh GC that provides a reduced nursing workload. To test the idea, data from 482 patients, with nine variables from four Malaysian intensive care units with different controls were exploited. The identified steps crucial in building a dependable model are variable selection, continuous state discretization, and unsupervised structure learning. Using a multi-target node evaluation, a network with 80{\%} mean overall classification precision was obtained with a normalized equal distance discretization algorithm and a maximum weight spanning tree technique. Meanwhile, the interest target node scored 90.39{\%} precision. The results from this study, which are complemented with an evaluation of missing data, are proposed as a benchmark for using Bayesian networks in this type of application.",
author = "Asma Abu-Samah and {Abdul Razak}, {Normy Norfiza} and Suhaimi, {Fatanah Mohamad} and Jamaludin, {Ummu Kulthum} and Chase, {James Geoffrey}",
year = "2019",
month = "1",
day = "1",
doi = "10.5573/IEIESPC.2019.8.3.202",
language = "English",
volume = "8",
pages = "202--209",
journal = "IEIE Transactions on Smart Processing and Computing",
issn = "2287-5255",
publisher = "Institute of Electronics and Information Engineers",
number = "3",

}

Towards personalized intensive care decision support using a Bayesian network : A multicenter glycemic control study. / Abu-Samah, Asma; Abdul Razak, Normy Norfiza; Suhaimi, Fatanah Mohamad; Jamaludin, Ummu Kulthum; Chase, James Geoffrey.

In: IEIE Transactions on Smart Processing and Computing, Vol. 8, No. 3, 01.01.2019, p. 202-209.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Towards personalized intensive care decision support using a Bayesian network

T2 - A multicenter glycemic control study

AU - Abu-Samah, Asma

AU - Abdul Razak, Normy Norfiza

AU - Suhaimi, Fatanah Mohamad

AU - Jamaludin, Ummu Kulthum

AU - Chase, James Geoffrey

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Personalized treatment in glycemic control (GC) is a visibly promising research area that requires improved mechanisms providing patient-specific procedures to enable complicated decision support. Available per-patient data must be more than written records, and be fully integrated in this personalization process. This article presents a process for relating the intensive care unit patients' demographic and admission data to their GC performance. With this objective, a probabilistic Bayesian network was chosen to provide more personalized decisions. As a case study, average daily blood glucose measurements were chosen as the interest target node in order to weigh GC that provides a reduced nursing workload. To test the idea, data from 482 patients, with nine variables from four Malaysian intensive care units with different controls were exploited. The identified steps crucial in building a dependable model are variable selection, continuous state discretization, and unsupervised structure learning. Using a multi-target node evaluation, a network with 80% mean overall classification precision was obtained with a normalized equal distance discretization algorithm and a maximum weight spanning tree technique. Meanwhile, the interest target node scored 90.39% precision. The results from this study, which are complemented with an evaluation of missing data, are proposed as a benchmark for using Bayesian networks in this type of application.

AB - Personalized treatment in glycemic control (GC) is a visibly promising research area that requires improved mechanisms providing patient-specific procedures to enable complicated decision support. Available per-patient data must be more than written records, and be fully integrated in this personalization process. This article presents a process for relating the intensive care unit patients' demographic and admission data to their GC performance. With this objective, a probabilistic Bayesian network was chosen to provide more personalized decisions. As a case study, average daily blood glucose measurements were chosen as the interest target node in order to weigh GC that provides a reduced nursing workload. To test the idea, data from 482 patients, with nine variables from four Malaysian intensive care units with different controls were exploited. The identified steps crucial in building a dependable model are variable selection, continuous state discretization, and unsupervised structure learning. Using a multi-target node evaluation, a network with 80% mean overall classification precision was obtained with a normalized equal distance discretization algorithm and a maximum weight spanning tree technique. Meanwhile, the interest target node scored 90.39% precision. The results from this study, which are complemented with an evaluation of missing data, are proposed as a benchmark for using Bayesian networks in this type of application.

UR - http://www.scopus.com/inward/record.url?scp=85068541334&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068541334&partnerID=8YFLogxK

U2 - 10.5573/IEIESPC.2019.8.3.202

DO - 10.5573/IEIESPC.2019.8.3.202

M3 - Article

VL - 8

SP - 202

EP - 209

JO - IEIE Transactions on Smart Processing and Computing

JF - IEIE Transactions on Smart Processing and Computing

SN - 2287-5255

IS - 3

ER -