Recognition of general anesthesia-induced loss of consciousness based on the spatial pattern of the brain networks

Objective. Unconsciousness is a key feature related to general anesthesia (GA) but is difficult to be evaluated accurately by anesthesiologists clinically. Approach. To tracking the loss of consciousness (LOC) and recovery of consciousness (ROC) under GA, in this study, by investigating functional connectivity of the scalp electroencephalogram, we explore any potential difference in brain networks among anesthesia induction, anesthesia recovery, and the resting state. Main results. The results of this study demonstrated significant differences among the three periods, concerning the corresponding brain networks. In detail, the suppressed default mode network, as well as the prolonged characteristic path length and decreased clustering coefficient, during LOC was found in the alpha band, compared to the Resting and the ROC state. When to further identify the Resting and LOC states, the fused network topologies and properties achieved the highest accuracy of 95%, along with a sensitivity of 93.33% and a specificity of 96.67%. Significance. The findings of this study not only deepen our understanding of propofol-induced unconsciousness but also provide quantitative measurements subserving better anesthesia management.

Automated summary

This study identified significant differences in brain networks during different periods of general anesthesia, with specific changes in the alpha band during LOC. The fused network topologies and properties achieved high accuracy in distinguishing between resting and LOC states, providing potential for better anesthesia management.