Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network

Glycemic control in critically ill patients is complex in terms of patients’ response to care and treatment. The variability and the search for improved insulin therapy outcomes have led to the use of human physiology model based on per-patient metabolic condition to provide automated recommendation...

Full description

Bibliographic Details
Main Authors:	Abu-Samah, Asma, Abdul Razak, Normy Norfiza, Suhaimi, Fatanah M., Jamaludin, Ummu Kulthum, Md Ralib, Azrina
Format:	Article
Language:	English English
Published:	Penerbit UTM Press 2019
Subjects:	R Medicine (General)
Online Access:	http://irep.iium.edu.my/68312/ http://irep.iium.edu.my/68312/ http://irep.iium.edu.my/68312/13/68312%20Probabilistic%20glycemic%20control%20decision%20support%20in%20ICU.pdf http://irep.iium.edu.my/68312/14/68312%20Probabilistic%20glycemic%20control%20decision%20support%20in%20ICU%20SCOPUS.pdf

id	iium-68312
recordtype	eprints
spelling	iium-683122019-08-02T01:59:08Z http://irep.iium.edu.my/68312/ Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network Abu-Samah, Asma Abdul Razak, Normy Norfiza Suhaimi, Fatanah M. Jamaludin, Ummu Kulthum Md Ralib, Azrina R Medicine (General) Glycemic control in critically ill patients is complex in terms of patients’ response to care and treatment. The variability and the search for improved insulin therapy outcomes have led to the use of human physiology model based on per-patient metabolic condition to provide automated recommendations. One of the most promising solution for this is the STAR protocol which is based on a clinically validated ICING insulin and nutrition physiological model, however this approach does not consider demographical background such as age, weight, height and ethnicity. This article presents the extension to their personalized care solution by integrating per-patient demographical and upon admission to intensive care conditions to automate decision support for clinical staffs. In this context, a virtual study was conducted on 210 retrospectives critically ill patients’ data. To provide a ground, the integration concept is presented roughly, but the details are given in terms of a proof of concept using Bayesian Network, linking the admission background and the STAR control’s performance. The proof of concept study shows the feasibility and the clinical potential to employ the probabilistic method as a decision support towards a more personalized care. ************************************************************************************* Kawalan glisemik dalam pesakit kritikal di unit rawatan rapi adalah rumit dari segi tindak balas pesakit terhadap penjagaan dan rawatan. Sifat keberubahan individu dan pencarian hasil terapi insulin yang lebih baik telah membawa kepada penggunaan model matematik fisiologi manusia berdasarkan keadaan metabolik pesakit untuk memberikan cadangan rawatan secara individu. Salah satu penyelesaian yang paling menjanjikan harapan adalah protokol STAR yang berdasarkan kepada model fisiologi insulin-nutrisi-glukosa yang telah disahkan secara klinikal. Namun pendekatan ini tidak mengambil kira latar belakang demografi seperti umur, berat, ketinggian dan etnik. Artikel ini membentangkan lanjutan kepada penyelesaian rawatan secara peribadi mereka dengan mengintegrasikan informasi demografi pesakit dan keadaan mereka semasa dimasukkan ke dalam unit rawatan rapi untuk mengautomasikan sokongan keputusan untuk kakitangan unit. Dalam konteks ini, satu kajian ‘virtual’ dilakukan pada data 210 pesaki. Sebagai kajian kes, konsep integrasi dibentangkan secara kasar, tetapi butiran diberikan dari segi bukti konsep yang menggunakan Rangkaian Bayesian, menghubungkan latar belakang kemasukan ke unit dan prestasi kawalan STAR. Bukti kajian kes menunjukkan 71.43% dan 73.90% ketepatan dan kebolehlaksanaan unjuran masing-masing dengan dataset ujian. Dengan lebih banyak data, rangkaian Bayesian yang lebih baik dipercayai boleh dihasilkan. Walaubagaimanapun, keputusan ini menunjukkan kemungkinan rangkaian ini bertindak sebagai pengelas yang berkesan dengan menggunakan data dari unit rawatan rapi dan prestasi kawalan glisemik untuk menjadi asas sokongan keputusan bersifat probabilistik, peribadi dan automatic dalam unit rawatan rapi. Penerbit UTM Press 2019-03 Article PeerReviewed application/pdf en http://irep.iium.edu.my/68312/13/68312%20Probabilistic%20glycemic%20control%20decision%20support%20in%20ICU.pdf application/pdf en http://irep.iium.edu.my/68312/14/68312%20Probabilistic%20glycemic%20control%20decision%20support%20in%20ICU%20SCOPUS.pdf Abu-Samah, Asma and Abdul Razak, Normy Norfiza and Suhaimi, Fatanah M. and Jamaludin, Ummu Kulthum and Md Ralib, Azrina (2019) Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network. Jurnal Teknologi (Sciences and Engineering), 81 (2). pp. 61-69. E-ISSN 2180-3772 https://jurnalteknologi.utm.my/index.php/jurnalteknologi/article/view/12721/6501
repository_type	Digital Repository
institution_category	Local University
institution	International Islamic University Malaysia
building	IIUM Repository
collection	Online Access
language	English English
topic	R Medicine (General)
spellingShingle	R Medicine (General) Abu-Samah, Asma Abdul Razak, Normy Norfiza Suhaimi, Fatanah M. Jamaludin, Ummu Kulthum Md Ralib, Azrina Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network
description	Glycemic control in critically ill patients is complex in terms of patients’ response to care and treatment. The variability and the search for improved insulin therapy outcomes have led to the use of human physiology model based on per-patient metabolic condition to provide automated recommendations. One of the most promising solution for this is the STAR protocol which is based on a clinically validated ICING insulin and nutrition physiological model, however this approach does not consider demographical background such as age, weight, height and ethnicity. This article presents the extension to their personalized care solution by integrating per-patient demographical and upon admission to intensive care conditions to automate decision support for clinical staffs. In this context, a virtual study was conducted on 210 retrospectives critically ill patients’ data. To provide a ground, the integration concept is presented roughly, but the details are given in terms of a proof of concept using Bayesian Network, linking the admission background and the STAR control’s performance. The proof of concept study shows the feasibility and the clinical potential to employ the probabilistic method as a decision support towards a more personalized care. ************************************************************************************* Kawalan glisemik dalam pesakit kritikal di unit rawatan rapi adalah rumit dari segi tindak balas pesakit terhadap penjagaan dan rawatan. Sifat keberubahan individu dan pencarian hasil terapi insulin yang lebih baik telah membawa kepada penggunaan model matematik fisiologi manusia berdasarkan keadaan metabolik pesakit untuk memberikan cadangan rawatan secara individu. Salah satu penyelesaian yang paling menjanjikan harapan adalah protokol STAR yang berdasarkan kepada model fisiologi insulin-nutrisi-glukosa yang telah disahkan secara klinikal. Namun pendekatan ini tidak mengambil kira latar belakang demografi seperti umur, berat, ketinggian dan etnik. Artikel ini membentangkan lanjutan kepada penyelesaian rawatan secara peribadi mereka dengan mengintegrasikan informasi demografi pesakit dan keadaan mereka semasa dimasukkan ke dalam unit rawatan rapi untuk mengautomasikan sokongan keputusan untuk kakitangan unit. Dalam konteks ini, satu kajian ‘virtual’ dilakukan pada data 210 pesaki. Sebagai kajian kes, konsep integrasi dibentangkan secara kasar, tetapi butiran diberikan dari segi bukti konsep yang menggunakan Rangkaian Bayesian, menghubungkan latar belakang kemasukan ke unit dan prestasi kawalan STAR. Bukti kajian kes menunjukkan 71.43% dan 73.90% ketepatan dan kebolehlaksanaan unjuran masing-masing dengan dataset ujian. Dengan lebih banyak data, rangkaian Bayesian yang lebih baik dipercayai boleh dihasilkan. Walaubagaimanapun, keputusan ini menunjukkan kemungkinan rangkaian ini bertindak sebagai pengelas yang berkesan dengan menggunakan data dari unit rawatan rapi dan prestasi kawalan glisemik untuk menjadi asas sokongan keputusan bersifat probabilistik, peribadi dan automatic dalam unit rawatan rapi.
format	Article
author	Abu-Samah, Asma Abdul Razak, Normy Norfiza Suhaimi, Fatanah M. Jamaludin, Ummu Kulthum Md Ralib, Azrina
author_facet	Abu-Samah, Asma Abdul Razak, Normy Norfiza Suhaimi, Fatanah M. Jamaludin, Ummu Kulthum Md Ralib, Azrina
author_sort	Abu-Samah, Asma
title	Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network
title_short	Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network
title_full	Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network
title_fullStr	Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network
title_full_unstemmed	Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network
title_sort	probabilistic glycemic control decision support in icu: proof of concept using bayesian network
publisher	Penerbit UTM Press
publishDate	2019
url	http://irep.iium.edu.my/68312/ http://irep.iium.edu.my/68312/ http://irep.iium.edu.my/68312/13/68312%20Probabilistic%20glycemic%20control%20decision%20support%20in%20ICU.pdf http://irep.iium.edu.my/68312/14/68312%20Probabilistic%20glycemic%20control%20decision%20support%20in%20ICU%20SCOPUS.pdf
first_indexed	2023-09-18T21:36:57Z
last_indexed	2023-09-18T21:36:57Z
_version_	1777412883890569216

Probabilistic glycemic control decision support in icu: proof of concept using Bayesian network

Similar Items