Pharmacologically informed machine learning approach for identifying pathological states of unconsciousness via resting-state fMRI

Justin M. Campbell, Zirui Huang, Jun Zhang, Xuehai Wu, Pengmin Qin, Georg Northoff, George A. Mashour, Anthony G. Hudetz

Research output: Contribution to journalArticle

Abstract

Determining the level of consciousness in patients with disorders of consciousness (DOC) remains challenging. To address this challenge, resting-state fMRI (rs-fMRI) has been widely used for detecting the local, regional, and network activity differences between DOC patients and healthy controls. Although substantial progress has been made towards this endeavor, the identification of robust rs-fMRI-based biomarkers for level of consciousness is still lacking. Recent developments in machine learning show promise as a tool to augment the discrimination between different states of consciousness in clinical practice. Here, we investigated whether machine learning models trained to make a binary distinction between conscious wakefulness and anesthetic-induced unconsciousness would then be capable of reliably identifying pathologically induced unconsciousness. We did so by extracting rs-fMRI-based features associated with local activity, regional homogeneity, and interregional functional activity in 44 subjects during wakefulness, light sedation, and unresponsiveness (deep sedation and general anesthesia), and subsequently using those features to train three distinct candidate machine learning classifiers: support vector machine, Extra Trees, artificial neural network. First, we show that all three classifiers achieve reliable performance within-dataset (via nested cross-validation), with a mean area under the receiver operating characteristic curve (AUC) of 0.95, 0.92, and 0.94, respectively. Additionally, we observed comparable cross-dataset performance (making predictions on the DOC data) as the anesthesia-trained classifiers demonstrated a consistent ability to discriminate between unresponsive wakefulness syndrome (UWS/VS) patients and healthy controls with mean AUC's of 0.99, 0.94, 0.98, respectively. Lastly, we explored the potential of applying the aforementioned classifiers towards discriminating intermediate states of consciousness, specifically, subjects under light anesthetic sedation and patients diagnosed as having a minimally conscious state (MCS). Our findings demonstrate that machine learning classifiers trained on rs-fMRI features derived from participants under anesthesia have potential to aid the discrimination between degrees of pathological unconsciousness in clinical patients.

Original languageEnglish
Article number116316
JournalNeuroImage
DOIs
Publication statusAccepted/In press - Jan 1 2019
Externally publishedYes

Fingerprint

Unconsciousness
Consciousness Disorders
Consciousness
Magnetic Resonance Imaging
Wakefulness
Area Under Curve
Anesthetics
Anesthesia
Deep Sedation
Persistent Vegetative State
Light
Aptitude
ROC Curve
General Anesthesia
Biomarkers
Machine Learning

Keywords

  • Anesthesia
  • Consciousness
  • Deep learning
  • Disorders of consciousness
  • fMRI
  • Functional connectivity
  • Machine learning
  • Resting-state

ASJC Scopus subject areas

  • Neurology
  • Cognitive Neuroscience

Cite this

Pharmacologically informed machine learning approach for identifying pathological states of unconsciousness via resting-state fMRI. / Campbell, Justin M.; Huang, Zirui; Zhang, Jun; Wu, Xuehai; Qin, Pengmin; Northoff, Georg; Mashour, George A.; Hudetz, Anthony G.

In: NeuroImage, 01.01.2019.

Research output: Contribution to journalArticle

Campbell, Justin M. ; Huang, Zirui ; Zhang, Jun ; Wu, Xuehai ; Qin, Pengmin ; Northoff, Georg ; Mashour, George A. ; Hudetz, Anthony G. / Pharmacologically informed machine learning approach for identifying pathological states of unconsciousness via resting-state fMRI. In: NeuroImage. 2019.
@article{658d6b2722f34584a12970afe354950c,
title = "Pharmacologically informed machine learning approach for identifying pathological states of unconsciousness via resting-state fMRI",
abstract = "Determining the level of consciousness in patients with disorders of consciousness (DOC) remains challenging. To address this challenge, resting-state fMRI (rs-fMRI) has been widely used for detecting the local, regional, and network activity differences between DOC patients and healthy controls. Although substantial progress has been made towards this endeavor, the identification of robust rs-fMRI-based biomarkers for level of consciousness is still lacking. Recent developments in machine learning show promise as a tool to augment the discrimination between different states of consciousness in clinical practice. Here, we investigated whether machine learning models trained to make a binary distinction between conscious wakefulness and anesthetic-induced unconsciousness would then be capable of reliably identifying pathologically induced unconsciousness. We did so by extracting rs-fMRI-based features associated with local activity, regional homogeneity, and interregional functional activity in 44 subjects during wakefulness, light sedation, and unresponsiveness (deep sedation and general anesthesia), and subsequently using those features to train three distinct candidate machine learning classifiers: support vector machine, Extra Trees, artificial neural network. First, we show that all three classifiers achieve reliable performance within-dataset (via nested cross-validation), with a mean area under the receiver operating characteristic curve (AUC) of 0.95, 0.92, and 0.94, respectively. Additionally, we observed comparable cross-dataset performance (making predictions on the DOC data) as the anesthesia-trained classifiers demonstrated a consistent ability to discriminate between unresponsive wakefulness syndrome (UWS/VS) patients and healthy controls with mean AUC's of 0.99, 0.94, 0.98, respectively. Lastly, we explored the potential of applying the aforementioned classifiers towards discriminating intermediate states of consciousness, specifically, subjects under light anesthetic sedation and patients diagnosed as having a minimally conscious state (MCS). Our findings demonstrate that machine learning classifiers trained on rs-fMRI features derived from participants under anesthesia have potential to aid the discrimination between degrees of pathological unconsciousness in clinical patients.",
keywords = "Anesthesia, Consciousness, Deep learning, Disorders of consciousness, fMRI, Functional connectivity, Machine learning, Resting-state",
author = "Campbell, {Justin M.} and Zirui Huang and Jun Zhang and Xuehai Wu and Pengmin Qin and Georg Northoff and Mashour, {George A.} and Hudetz, {Anthony G.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.neuroimage.2019.116316",
language = "English",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Pharmacologically informed machine learning approach for identifying pathological states of unconsciousness via resting-state fMRI

AU - Campbell, Justin M.

AU - Huang, Zirui

AU - Zhang, Jun

AU - Wu, Xuehai

AU - Qin, Pengmin

AU - Northoff, Georg

AU - Mashour, George A.

AU - Hudetz, Anthony G.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Determining the level of consciousness in patients with disorders of consciousness (DOC) remains challenging. To address this challenge, resting-state fMRI (rs-fMRI) has been widely used for detecting the local, regional, and network activity differences between DOC patients and healthy controls. Although substantial progress has been made towards this endeavor, the identification of robust rs-fMRI-based biomarkers for level of consciousness is still lacking. Recent developments in machine learning show promise as a tool to augment the discrimination between different states of consciousness in clinical practice. Here, we investigated whether machine learning models trained to make a binary distinction between conscious wakefulness and anesthetic-induced unconsciousness would then be capable of reliably identifying pathologically induced unconsciousness. We did so by extracting rs-fMRI-based features associated with local activity, regional homogeneity, and interregional functional activity in 44 subjects during wakefulness, light sedation, and unresponsiveness (deep sedation and general anesthesia), and subsequently using those features to train three distinct candidate machine learning classifiers: support vector machine, Extra Trees, artificial neural network. First, we show that all three classifiers achieve reliable performance within-dataset (via nested cross-validation), with a mean area under the receiver operating characteristic curve (AUC) of 0.95, 0.92, and 0.94, respectively. Additionally, we observed comparable cross-dataset performance (making predictions on the DOC data) as the anesthesia-trained classifiers demonstrated a consistent ability to discriminate between unresponsive wakefulness syndrome (UWS/VS) patients and healthy controls with mean AUC's of 0.99, 0.94, 0.98, respectively. Lastly, we explored the potential of applying the aforementioned classifiers towards discriminating intermediate states of consciousness, specifically, subjects under light anesthetic sedation and patients diagnosed as having a minimally conscious state (MCS). Our findings demonstrate that machine learning classifiers trained on rs-fMRI features derived from participants under anesthesia have potential to aid the discrimination between degrees of pathological unconsciousness in clinical patients.

AB - Determining the level of consciousness in patients with disorders of consciousness (DOC) remains challenging. To address this challenge, resting-state fMRI (rs-fMRI) has been widely used for detecting the local, regional, and network activity differences between DOC patients and healthy controls. Although substantial progress has been made towards this endeavor, the identification of robust rs-fMRI-based biomarkers for level of consciousness is still lacking. Recent developments in machine learning show promise as a tool to augment the discrimination between different states of consciousness in clinical practice. Here, we investigated whether machine learning models trained to make a binary distinction between conscious wakefulness and anesthetic-induced unconsciousness would then be capable of reliably identifying pathologically induced unconsciousness. We did so by extracting rs-fMRI-based features associated with local activity, regional homogeneity, and interregional functional activity in 44 subjects during wakefulness, light sedation, and unresponsiveness (deep sedation and general anesthesia), and subsequently using those features to train three distinct candidate machine learning classifiers: support vector machine, Extra Trees, artificial neural network. First, we show that all three classifiers achieve reliable performance within-dataset (via nested cross-validation), with a mean area under the receiver operating characteristic curve (AUC) of 0.95, 0.92, and 0.94, respectively. Additionally, we observed comparable cross-dataset performance (making predictions on the DOC data) as the anesthesia-trained classifiers demonstrated a consistent ability to discriminate between unresponsive wakefulness syndrome (UWS/VS) patients and healthy controls with mean AUC's of 0.99, 0.94, 0.98, respectively. Lastly, we explored the potential of applying the aforementioned classifiers towards discriminating intermediate states of consciousness, specifically, subjects under light anesthetic sedation and patients diagnosed as having a minimally conscious state (MCS). Our findings demonstrate that machine learning classifiers trained on rs-fMRI features derived from participants under anesthesia have potential to aid the discrimination between degrees of pathological unconsciousness in clinical patients.

KW - Anesthesia

KW - Consciousness

KW - Deep learning

KW - Disorders of consciousness

KW - fMRI

KW - Functional connectivity

KW - Machine learning

KW - Resting-state

UR - http://www.scopus.com/inward/record.url?scp=85075482968&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075482968&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2019.116316

DO - 10.1016/j.neuroimage.2019.116316

M3 - Article

C2 - 31672663

AN - SCOPUS:85075482968

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

M1 - 116316

ER -