Estimation of caffeine regimens: A machine learning approach for enhanced clinical decision making at a neonatal intensive care unit (NICU)

Rudresh Deepak Shirwaikar, U. Dinesh Acharya, Krishnamoorthi Makkithaya, Surulivelrajan Mallayaswamy, Leslie Edward Simon Lewis

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The decision-making process for estimating the optimal dosage is critical in clinical settings. In the neonatal intensive care unit (NICU), preterm neonates suffering from apnea of prematurity, optimum drug dosage can make a difference between life and death. To improve clinical decision making in the NICU, we have developed prediction models using machine learning algorithms. We have used optimized Support Vector Machine (SVM), decision trees with ensembles created using Bagging, Boosting, Random Forest, optimized Multi Layer Perceptron (MLP) and Deep Learning to predict adequacy of caffeine, a methylxanthine used to prevent the development of recurrent apneas, to reduce the need for mechanical ventilation. The respective models developed were evaluated using 100 clinical caffeine cases collected from the Neonatal Intensive Care Unit (NICU) of Kasturba Medical College, Manipal. Our results indicate that a deep belief network (DBN) having an area under curve (AUC) of 0.91, followed by an optimized MLP with the Score for Neonatal Acute Physiology I (SNAP I) as an input feature, outperform other models for assessing the drug effectiveness. Furthermore, the optimized MLP followed by a DBN, with SNAP I as an input feature is a more accurate model for predicting the therapeutic concentration of caffeine. These results suggest that the proposed SNAP I (illness severity score) acts as a critical input variable to enhance the performance of the prediction model. The machine learning approach is very useful for building decision support systems in the NICU in general, and it provides specific solutions to optimize the administration of lifesaving drugs to neonates who are very sensitive to dosages. Using our method, physicians can assess the adequacy and efficacy of caffeine on the study population in a NICU before administering it to neonates.

Original languageEnglish
Pages (from-to)93-115
Number of pages23
JournalCritical Reviews in Biomedical Engineering
Volume46
Issue number2
DOIs
Publication statusPublished - 01-01-2018

Fingerprint

Caffeine
Intensive care units
Learning systems
Decision making
Physiology
Multilayer neural networks
Bayesian networks
Drug dosage
Decision trees
Decision support systems
Learning algorithms
Support vector machines

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering

Cite this

@article{53402fd12332418a87c6641aa32205b9,
title = "Estimation of caffeine regimens: A machine learning approach for enhanced clinical decision making at a neonatal intensive care unit (NICU)",
abstract = "The decision-making process for estimating the optimal dosage is critical in clinical settings. In the neonatal intensive care unit (NICU), preterm neonates suffering from apnea of prematurity, optimum drug dosage can make a difference between life and death. To improve clinical decision making in the NICU, we have developed prediction models using machine learning algorithms. We have used optimized Support Vector Machine (SVM), decision trees with ensembles created using Bagging, Boosting, Random Forest, optimized Multi Layer Perceptron (MLP) and Deep Learning to predict adequacy of caffeine, a methylxanthine used to prevent the development of recurrent apneas, to reduce the need for mechanical ventilation. The respective models developed were evaluated using 100 clinical caffeine cases collected from the Neonatal Intensive Care Unit (NICU) of Kasturba Medical College, Manipal. Our results indicate that a deep belief network (DBN) having an area under curve (AUC) of 0.91, followed by an optimized MLP with the Score for Neonatal Acute Physiology I (SNAP I) as an input feature, outperform other models for assessing the drug effectiveness. Furthermore, the optimized MLP followed by a DBN, with SNAP I as an input feature is a more accurate model for predicting the therapeutic concentration of caffeine. These results suggest that the proposed SNAP I (illness severity score) acts as a critical input variable to enhance the performance of the prediction model. The machine learning approach is very useful for building decision support systems in the NICU in general, and it provides specific solutions to optimize the administration of lifesaving drugs to neonates who are very sensitive to dosages. Using our method, physicians can assess the adequacy and efficacy of caffeine on the study population in a NICU before administering it to neonates.",
author = "Shirwaikar, {Rudresh Deepak} and {Dinesh Acharya}, U. and Krishnamoorthi Makkithaya and Surulivelrajan Mallayaswamy and Lewis, {Leslie Edward Simon}",
year = "2018",
month = "1",
day = "1",
doi = "10.1615/CritRevBiomedEng.2018025933",
language = "English",
volume = "46",
pages = "93--115",
journal = "Critical Reviews in Biomedical Engineering",
issn = "0278-940X",
publisher = "Begell House Inc.",
number = "2",

}

TY - JOUR

T1 - Estimation of caffeine regimens

T2 - A machine learning approach for enhanced clinical decision making at a neonatal intensive care unit (NICU)

AU - Shirwaikar, Rudresh Deepak

AU - Dinesh Acharya, U.

AU - Makkithaya, Krishnamoorthi

AU - Mallayaswamy, Surulivelrajan

AU - Lewis, Leslie Edward Simon

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The decision-making process for estimating the optimal dosage is critical in clinical settings. In the neonatal intensive care unit (NICU), preterm neonates suffering from apnea of prematurity, optimum drug dosage can make a difference between life and death. To improve clinical decision making in the NICU, we have developed prediction models using machine learning algorithms. We have used optimized Support Vector Machine (SVM), decision trees with ensembles created using Bagging, Boosting, Random Forest, optimized Multi Layer Perceptron (MLP) and Deep Learning to predict adequacy of caffeine, a methylxanthine used to prevent the development of recurrent apneas, to reduce the need for mechanical ventilation. The respective models developed were evaluated using 100 clinical caffeine cases collected from the Neonatal Intensive Care Unit (NICU) of Kasturba Medical College, Manipal. Our results indicate that a deep belief network (DBN) having an area under curve (AUC) of 0.91, followed by an optimized MLP with the Score for Neonatal Acute Physiology I (SNAP I) as an input feature, outperform other models for assessing the drug effectiveness. Furthermore, the optimized MLP followed by a DBN, with SNAP I as an input feature is a more accurate model for predicting the therapeutic concentration of caffeine. These results suggest that the proposed SNAP I (illness severity score) acts as a critical input variable to enhance the performance of the prediction model. The machine learning approach is very useful for building decision support systems in the NICU in general, and it provides specific solutions to optimize the administration of lifesaving drugs to neonates who are very sensitive to dosages. Using our method, physicians can assess the adequacy and efficacy of caffeine on the study population in a NICU before administering it to neonates.

AB - The decision-making process for estimating the optimal dosage is critical in clinical settings. In the neonatal intensive care unit (NICU), preterm neonates suffering from apnea of prematurity, optimum drug dosage can make a difference between life and death. To improve clinical decision making in the NICU, we have developed prediction models using machine learning algorithms. We have used optimized Support Vector Machine (SVM), decision trees with ensembles created using Bagging, Boosting, Random Forest, optimized Multi Layer Perceptron (MLP) and Deep Learning to predict adequacy of caffeine, a methylxanthine used to prevent the development of recurrent apneas, to reduce the need for mechanical ventilation. The respective models developed were evaluated using 100 clinical caffeine cases collected from the Neonatal Intensive Care Unit (NICU) of Kasturba Medical College, Manipal. Our results indicate that a deep belief network (DBN) having an area under curve (AUC) of 0.91, followed by an optimized MLP with the Score for Neonatal Acute Physiology I (SNAP I) as an input feature, outperform other models for assessing the drug effectiveness. Furthermore, the optimized MLP followed by a DBN, with SNAP I as an input feature is a more accurate model for predicting the therapeutic concentration of caffeine. These results suggest that the proposed SNAP I (illness severity score) acts as a critical input variable to enhance the performance of the prediction model. The machine learning approach is very useful for building decision support systems in the NICU in general, and it provides specific solutions to optimize the administration of lifesaving drugs to neonates who are very sensitive to dosages. Using our method, physicians can assess the adequacy and efficacy of caffeine on the study population in a NICU before administering it to neonates.

UR - http://www.scopus.com/inward/record.url?scp=85055329226&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055329226&partnerID=8YFLogxK

U2 - 10.1615/CritRevBiomedEng.2018025933

DO - 10.1615/CritRevBiomedEng.2018025933

M3 - Article

AN - SCOPUS:85055329226

VL - 46

SP - 93

EP - 115

JO - Critical Reviews in Biomedical Engineering

JF - Critical Reviews in Biomedical Engineering

SN - 0278-940X

IS - 2

ER -