Applying artificial intelligence to identify physiomarkers predicting severe sepsis in the PICU

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Objectives: We used artificial intelligence to develop a novel algorithm using physiomarkers to predict the onset of severe sepsis in critically ill children. Design: Observational cohort study. Setting: PICU. Patients: Children age between 6 and 18 years old. Interventions: None. Measurements and Main Results: Continuous minute-by-minute physiologic data were available for a total of 493 critically ill children admitted to a tertiary care PICU over an 8-month period, 20 of whom developed severe sepsis. Using an alert time stamp generated by an electronic screening algorithm as a reference point, we studied up to 24 prior hours of continuous physiologic data. We identified physiomarkers, including sd of heart rate, systolic and diastolic blood pressure, and symbolic transitions probabilities of those variables that discriminated severe sepsis patients from controls (all other patients admitted to the PICU who did not meet severe sepsis criteria). We used logistic regression, random forests, and deep Convolutional Neural Network methods to derive our models. Analysis was performed using data generated in two windows prior to the firing of the electronic screening algorithm, namely, 2-8 and 8-24 hours. When analyzing the physiomarkers present in the 2-8 hours analysis window, logistic regression performed with specificity of 87.4% and sensitivity of 55.0%, random forest performed with 79.6% specificity and 80.0% sensitivity, and the Convolutional Neural Network performed with 83.0% specificity and 75.0% sensitivity. When analyzing physiomarkers from the 8-24 hours window, logistic regression resulted in 77.1% specificity and 39.3% sensitivity, random forest performed with 82.3% specificity and 61.1% sensitivity, whereas the Convolutional Neural Network method achieved 81% specificity and 76% sensitivity. Conclusions: Artificial intelligence can be used to predict the onset of severe sepsis using physiomarkers in critically ill children. Further, it may detect severe sepsis as early as 8 hours prior to a real-time electronic severe sepsis screening algorithm.

Original languageEnglish (US)
Pages (from-to)E495-E503
JournalPediatric Critical Care Medicine
Volume19
Issue number10
DOIs
StatePublished - Jan 1 2018

Fingerprint

Artificial Intelligence
Sepsis
Sensitivity and Specificity
Critical Illness
Logistic Models
Blood Pressure
Tertiary Healthcare
Observational Studies
Cohort Studies
Heart Rate
Forests

All Science Journal Classification (ASJC) codes

  • Pediatrics, Perinatology, and Child Health
  • Critical Care and Intensive Care Medicine

Cite this

@article{c73fe94b799246a0bbf12c94d032dc2b,
title = "Applying artificial intelligence to identify physiomarkers predicting severe sepsis in the PICU",
abstract = "Objectives: We used artificial intelligence to develop a novel algorithm using physiomarkers to predict the onset of severe sepsis in critically ill children. Design: Observational cohort study. Setting: PICU. Patients: Children age between 6 and 18 years old. Interventions: None. Measurements and Main Results: Continuous minute-by-minute physiologic data were available for a total of 493 critically ill children admitted to a tertiary care PICU over an 8-month period, 20 of whom developed severe sepsis. Using an alert time stamp generated by an electronic screening algorithm as a reference point, we studied up to 24 prior hours of continuous physiologic data. We identified physiomarkers, including sd of heart rate, systolic and diastolic blood pressure, and symbolic transitions probabilities of those variables that discriminated severe sepsis patients from controls (all other patients admitted to the PICU who did not meet severe sepsis criteria). We used logistic regression, random forests, and deep Convolutional Neural Network methods to derive our models. Analysis was performed using data generated in two windows prior to the firing of the electronic screening algorithm, namely, 2-8 and 8-24 hours. When analyzing the physiomarkers present in the 2-8 hours analysis window, logistic regression performed with specificity of 87.4{\%} and sensitivity of 55.0{\%}, random forest performed with 79.6{\%} specificity and 80.0{\%} sensitivity, and the Convolutional Neural Network performed with 83.0{\%} specificity and 75.0{\%} sensitivity. When analyzing physiomarkers from the 8-24 hours window, logistic regression resulted in 77.1{\%} specificity and 39.3{\%} sensitivity, random forest performed with 82.3{\%} specificity and 61.1{\%} sensitivity, whereas the Convolutional Neural Network method achieved 81{\%} specificity and 76{\%} sensitivity. Conclusions: Artificial intelligence can be used to predict the onset of severe sepsis using physiomarkers in critically ill children. Further, it may detect severe sepsis as early as 8 hours prior to a real-time electronic severe sepsis screening algorithm.",
author = "Rishikesan Kamaleswaran and Oguz Akbilgic and Hallman, {Madhura A.} and Alina West and Robert Davis and Samir Shah",
year = "2018",
month = "1",
day = "1",
doi = "10.1097/PCC.0000000000001666",
language = "English (US)",
volume = "19",
pages = "E495--E503",
journal = "Pediatric Critical Care Medicine",
issn = "1529-7535",
publisher = "Lippincott Williams and Wilkins",
number = "10",

}

TY - JOUR

T1 - Applying artificial intelligence to identify physiomarkers predicting severe sepsis in the PICU

AU - Kamaleswaran, Rishikesan

AU - Akbilgic, Oguz

AU - Hallman, Madhura A.

AU - West, Alina

AU - Davis, Robert

AU - Shah, Samir

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Objectives: We used artificial intelligence to develop a novel algorithm using physiomarkers to predict the onset of severe sepsis in critically ill children. Design: Observational cohort study. Setting: PICU. Patients: Children age between 6 and 18 years old. Interventions: None. Measurements and Main Results: Continuous minute-by-minute physiologic data were available for a total of 493 critically ill children admitted to a tertiary care PICU over an 8-month period, 20 of whom developed severe sepsis. Using an alert time stamp generated by an electronic screening algorithm as a reference point, we studied up to 24 prior hours of continuous physiologic data. We identified physiomarkers, including sd of heart rate, systolic and diastolic blood pressure, and symbolic transitions probabilities of those variables that discriminated severe sepsis patients from controls (all other patients admitted to the PICU who did not meet severe sepsis criteria). We used logistic regression, random forests, and deep Convolutional Neural Network methods to derive our models. Analysis was performed using data generated in two windows prior to the firing of the electronic screening algorithm, namely, 2-8 and 8-24 hours. When analyzing the physiomarkers present in the 2-8 hours analysis window, logistic regression performed with specificity of 87.4% and sensitivity of 55.0%, random forest performed with 79.6% specificity and 80.0% sensitivity, and the Convolutional Neural Network performed with 83.0% specificity and 75.0% sensitivity. When analyzing physiomarkers from the 8-24 hours window, logistic regression resulted in 77.1% specificity and 39.3% sensitivity, random forest performed with 82.3% specificity and 61.1% sensitivity, whereas the Convolutional Neural Network method achieved 81% specificity and 76% sensitivity. Conclusions: Artificial intelligence can be used to predict the onset of severe sepsis using physiomarkers in critically ill children. Further, it may detect severe sepsis as early as 8 hours prior to a real-time electronic severe sepsis screening algorithm.

AB - Objectives: We used artificial intelligence to develop a novel algorithm using physiomarkers to predict the onset of severe sepsis in critically ill children. Design: Observational cohort study. Setting: PICU. Patients: Children age between 6 and 18 years old. Interventions: None. Measurements and Main Results: Continuous minute-by-minute physiologic data were available for a total of 493 critically ill children admitted to a tertiary care PICU over an 8-month period, 20 of whom developed severe sepsis. Using an alert time stamp generated by an electronic screening algorithm as a reference point, we studied up to 24 prior hours of continuous physiologic data. We identified physiomarkers, including sd of heart rate, systolic and diastolic blood pressure, and symbolic transitions probabilities of those variables that discriminated severe sepsis patients from controls (all other patients admitted to the PICU who did not meet severe sepsis criteria). We used logistic regression, random forests, and deep Convolutional Neural Network methods to derive our models. Analysis was performed using data generated in two windows prior to the firing of the electronic screening algorithm, namely, 2-8 and 8-24 hours. When analyzing the physiomarkers present in the 2-8 hours analysis window, logistic regression performed with specificity of 87.4% and sensitivity of 55.0%, random forest performed with 79.6% specificity and 80.0% sensitivity, and the Convolutional Neural Network performed with 83.0% specificity and 75.0% sensitivity. When analyzing physiomarkers from the 8-24 hours window, logistic regression resulted in 77.1% specificity and 39.3% sensitivity, random forest performed with 82.3% specificity and 61.1% sensitivity, whereas the Convolutional Neural Network method achieved 81% specificity and 76% sensitivity. Conclusions: Artificial intelligence can be used to predict the onset of severe sepsis using physiomarkers in critically ill children. Further, it may detect severe sepsis as early as 8 hours prior to a real-time electronic severe sepsis screening algorithm.

UR - http://www.scopus.com/inward/record.url?scp=85054397306&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054397306&partnerID=8YFLogxK

U2 - 10.1097/PCC.0000000000001666

DO - 10.1097/PCC.0000000000001666

M3 - Article

VL - 19

SP - E495-E503

JO - Pediatric Critical Care Medicine

JF - Pediatric Critical Care Medicine

SN - 1529-7535

IS - 10

ER -