Investigating dynamic functional network patterns after propofol-induced loss of consciousness

We investigated the changes of dynamic brain functional network from awaken state to the anesthesia level suitable for surgery.

60-channel EEG data of 22 subjects are acquired at wakefulness, light anesthesia and deep anesthesia. The activity of 68 cortical regions are obtained by using EEG source imaging. Sliding window analysis is employed to obtain a dynamic sequence of brain functional network. K-means clustering algorithm is then employed to identify the common brain functional network patterns.

Five common brain functional network patterns were identified across all conscious levels. The occurrence of each meta-stable network pattern was associated with the level of anesthesia. A transition functional network pattern was found to transfer to the anesthesia dominating or wakefulness dominating network pattern depending on the conscious level. Furthermore, a functional network pattern persisted during both wakefulness and anesthesia is found to be supported by the anatomical connectivity.

Dynamic changes of brain functional network exist in both awaken and anesthesia state.

These findings suggest that dynamic brain functional network analysis plays a critical role in decoding the mechanism of general anesthesia. The obtained five metastable network patterns may be employed for monitoring the depth of anesthesia.