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

• Published in: NeuroImage (Orlando, Fla.) Vol. 206; p. 116316
• Main Authors: Campbell, Justin M., Huang, Zirui, Zhang, Jun, Wu, Xuehai, Qin, Pengmin, Northoff, Georg, Mashour, George A., Hudetz, Anthony G.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.02.2020; Elsevier Limited; Elsevier
• Subjects: Adolescent; Adult; Aged; Agonists; Anesthesia; Anesthesia, General; Brain - diagnostic imaging; Brain - physiopathology; Brain research; Child; Conscious Sedation; Consciousness; Consciousness Disorders - diagnostic imaging; Consciousness Disorders - physiopathology; Datasets; Deep learning; Deep Sedation; Disorders of consciousness; Female; fMRI; Functional connectivity; Functional magnetic resonance imaging; Functional Neuroimaging; Gas flow; Humans; Learning algorithms; Machine Learning; Magnetic Resonance Imaging; Male; Medical imaging; Medical personnel; Middle Aged; Neural networks; Neural Networks, Computer; Persistent Vegetative State - diagnostic imaging; Persistent Vegetative State - physiopathology; Rest; Resting-state; Sleep and wakefulness; Studies; Support Vector Machine; Unconsciousness; Unconsciousness - diagnostic imaging; Unconsciousness - physiopathology; Wakefulness; Young Adult; United States > US; Wisconsin; Functional connectivity; Deep learning; fMRI; Consciousness; Disorders of consciousness; Machine learning; Resting-state; Anesthesia
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2019.116316

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. •Anesthetic-induced unconsciousness informs pathological unconsciousness.•Rs-fMRI-based machine learning is applied to distinguish conscious from unconscious subjects.•Machine learning models achieve reliable performance in cross-dataset generalizations.•Hyperparameter optimization significantly improves model performance.•Network-level functional connectivity features are most informative for classification.