Creation of a Robust and Generalizable Machine Learning Classifier for Patient Ventilator Asynchrony

Gregory B. Rehm, Jinyoung Han, Brooks Kuhn, Jean Pierre Delplanque, Nicholas Anderson, Jason Yeates Adams, Chen Nee Chuah

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Summary Background: As healthcare increasingly digitizes, streaming waveform data is being made from an variety of sources, but there still remains a paucity of performant clinical decision support systems. For example, in the intensive care unit (ICU) existing automated alarm systems typically rely on simple thresholding that result in frequent false positives. Recurrent false positive alerts create distrust of alarm mechanisms that can be directly detrimental to patient health. To improve patient care in the ICU, we need alert systems that are both pervasive, and accurate so as to be informative and trusted by providers. Objective: We aimed to develop a machine learning-based classifier to detect abnormal waveform events using the use case of mechanical ventilation waveform analysis, and the detection of harmful forms of ventilation delivery to patients. We specifically focused on detecting injurious subtypes of patient-ventilator asynchrony (PVA). Methods: Using a dataset of breaths recorded from 35 different patients, we used machine learning to create computational models to automatically detect, and classify two types of injurious PVA, double trigger asynchrony (DTA), breath stacking asynchrony (BSA). We examined the use of synthetic minority over-sampling technique (SMOTE) to overcome class imbalance problems, varied methods for feature selection, and use of ensemble methods to optimize the performance of our model. Results: We created an ensemble classifier that is able to accurately detect DTA at a sensitivity/specificity of 0.960/0.975, BSA at sensitivity/specificity of 0.944/0.987, and non-PVA events at sensitivity/specificity of.967/.980. Conclusions: Our results suggest that it is possible to create a high-performing machine learning-based model for detecting PVA in mechanical ventilator waveform data in spite of both intra-patient, and inter-patient variability in waveform patterns, and the presence of clinical artifacts like cough and suction procedures. Our work highlights the importance of addressing class imbalance in clinical data sets, and the combined use of statistical methods and expert knowledge in feature selection.

Original languageEnglish (US)
Pages (from-to)208-219
Number of pages12
JournalMethods of Information in Medicine
Volume57
Issue number4
DOIs
StatePublished - Jan 1 2018

Fingerprint

Mechanical Ventilators
Sensitivity and Specificity
Intensive Care Units
Clinical Decision Support Systems
Machine Learning
Suction
Artificial Respiration
Cough
Artifacts
Ventilation
Patient Care
Delivery of Health Care
Health

Keywords

  • artificial respiration
  • clinical decision support systems
  • intensive care units
  • machine learning
  • Patient ventilator asynchrony

ASJC Scopus subject areas

  • Health Informatics
  • Advanced and Specialized Nursing
  • Health Information Management

Cite this

Creation of a Robust and Generalizable Machine Learning Classifier for Patient Ventilator Asynchrony. / Rehm, Gregory B.; Han, Jinyoung; Kuhn, Brooks; Delplanque, Jean Pierre; Anderson, Nicholas; Adams, Jason Yeates; Chuah, Chen Nee.

In: Methods of Information in Medicine, Vol. 57, No. 4, 01.01.2018, p. 208-219.

Research output: Contribution to journalArticle

@article{1991e10c706c4b018310107c9a62e2a2,
title = "Creation of a Robust and Generalizable Machine Learning Classifier for Patient Ventilator Asynchrony",
abstract = "Summary Background: As healthcare increasingly digitizes, streaming waveform data is being made from an variety of sources, but there still remains a paucity of performant clinical decision support systems. For example, in the intensive care unit (ICU) existing automated alarm systems typically rely on simple thresholding that result in frequent false positives. Recurrent false positive alerts create distrust of alarm mechanisms that can be directly detrimental to patient health. To improve patient care in the ICU, we need alert systems that are both pervasive, and accurate so as to be informative and trusted by providers. Objective: We aimed to develop a machine learning-based classifier to detect abnormal waveform events using the use case of mechanical ventilation waveform analysis, and the detection of harmful forms of ventilation delivery to patients. We specifically focused on detecting injurious subtypes of patient-ventilator asynchrony (PVA). Methods: Using a dataset of breaths recorded from 35 different patients, we used machine learning to create computational models to automatically detect, and classify two types of injurious PVA, double trigger asynchrony (DTA), breath stacking asynchrony (BSA). We examined the use of synthetic minority over-sampling technique (SMOTE) to overcome class imbalance problems, varied methods for feature selection, and use of ensemble methods to optimize the performance of our model. Results: We created an ensemble classifier that is able to accurately detect DTA at a sensitivity/specificity of 0.960/0.975, BSA at sensitivity/specificity of 0.944/0.987, and non-PVA events at sensitivity/specificity of.967/.980. Conclusions: Our results suggest that it is possible to create a high-performing machine learning-based model for detecting PVA in mechanical ventilator waveform data in spite of both intra-patient, and inter-patient variability in waveform patterns, and the presence of clinical artifacts like cough and suction procedures. Our work highlights the importance of addressing class imbalance in clinical data sets, and the combined use of statistical methods and expert knowledge in feature selection.",
keywords = "artificial respiration, clinical decision support systems, intensive care units, machine learning, Patient ventilator asynchrony",
author = "Rehm, {Gregory B.} and Jinyoung Han and Brooks Kuhn and Delplanque, {Jean Pierre} and Nicholas Anderson and Adams, {Jason Yeates} and Chuah, {Chen Nee}",
year = "2018",
month = "1",
day = "1",
doi = "10.3414/ME17-02-0012",
language = "English (US)",
volume = "57",
pages = "208--219",
journal = "Methods of Information in Medicine",
issn = "0026-1270",
publisher = "Schattauer GmbH",
number = "4",

}

TY - JOUR

T1 - Creation of a Robust and Generalizable Machine Learning Classifier for Patient Ventilator Asynchrony

AU - Rehm, Gregory B.

AU - Han, Jinyoung

AU - Kuhn, Brooks

AU - Delplanque, Jean Pierre

AU - Anderson, Nicholas

AU - Adams, Jason Yeates

AU - Chuah, Chen Nee

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Summary Background: As healthcare increasingly digitizes, streaming waveform data is being made from an variety of sources, but there still remains a paucity of performant clinical decision support systems. For example, in the intensive care unit (ICU) existing automated alarm systems typically rely on simple thresholding that result in frequent false positives. Recurrent false positive alerts create distrust of alarm mechanisms that can be directly detrimental to patient health. To improve patient care in the ICU, we need alert systems that are both pervasive, and accurate so as to be informative and trusted by providers. Objective: We aimed to develop a machine learning-based classifier to detect abnormal waveform events using the use case of mechanical ventilation waveform analysis, and the detection of harmful forms of ventilation delivery to patients. We specifically focused on detecting injurious subtypes of patient-ventilator asynchrony (PVA). Methods: Using a dataset of breaths recorded from 35 different patients, we used machine learning to create computational models to automatically detect, and classify two types of injurious PVA, double trigger asynchrony (DTA), breath stacking asynchrony (BSA). We examined the use of synthetic minority over-sampling technique (SMOTE) to overcome class imbalance problems, varied methods for feature selection, and use of ensemble methods to optimize the performance of our model. Results: We created an ensemble classifier that is able to accurately detect DTA at a sensitivity/specificity of 0.960/0.975, BSA at sensitivity/specificity of 0.944/0.987, and non-PVA events at sensitivity/specificity of.967/.980. Conclusions: Our results suggest that it is possible to create a high-performing machine learning-based model for detecting PVA in mechanical ventilator waveform data in spite of both intra-patient, and inter-patient variability in waveform patterns, and the presence of clinical artifacts like cough and suction procedures. Our work highlights the importance of addressing class imbalance in clinical data sets, and the combined use of statistical methods and expert knowledge in feature selection.

AB - Summary Background: As healthcare increasingly digitizes, streaming waveform data is being made from an variety of sources, but there still remains a paucity of performant clinical decision support systems. For example, in the intensive care unit (ICU) existing automated alarm systems typically rely on simple thresholding that result in frequent false positives. Recurrent false positive alerts create distrust of alarm mechanisms that can be directly detrimental to patient health. To improve patient care in the ICU, we need alert systems that are both pervasive, and accurate so as to be informative and trusted by providers. Objective: We aimed to develop a machine learning-based classifier to detect abnormal waveform events using the use case of mechanical ventilation waveform analysis, and the detection of harmful forms of ventilation delivery to patients. We specifically focused on detecting injurious subtypes of patient-ventilator asynchrony (PVA). Methods: Using a dataset of breaths recorded from 35 different patients, we used machine learning to create computational models to automatically detect, and classify two types of injurious PVA, double trigger asynchrony (DTA), breath stacking asynchrony (BSA). We examined the use of synthetic minority over-sampling technique (SMOTE) to overcome class imbalance problems, varied methods for feature selection, and use of ensemble methods to optimize the performance of our model. Results: We created an ensemble classifier that is able to accurately detect DTA at a sensitivity/specificity of 0.960/0.975, BSA at sensitivity/specificity of 0.944/0.987, and non-PVA events at sensitivity/specificity of.967/.980. Conclusions: Our results suggest that it is possible to create a high-performing machine learning-based model for detecting PVA in mechanical ventilator waveform data in spite of both intra-patient, and inter-patient variability in waveform patterns, and the presence of clinical artifacts like cough and suction procedures. Our work highlights the importance of addressing class imbalance in clinical data sets, and the combined use of statistical methods and expert knowledge in feature selection.

KW - artificial respiration

KW - clinical decision support systems

KW - intensive care units

KW - machine learning

KW - Patient ventilator asynchrony

UR - http://www.scopus.com/inward/record.url?scp=85053929350&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053929350&partnerID=8YFLogxK

U2 - 10.3414/ME17-02-0012

DO - 10.3414/ME17-02-0012

M3 - Article

VL - 57

SP - 208

EP - 219

JO - Methods of Information in Medicine

JF - Methods of Information in Medicine

SN - 0026-1270

IS - 4

ER -