EEG Based Brain Functional Analysis On Sedation Induction Using Transcutaneous Acupoint Electrical Stimulation And Propofol

In clinical practice,the dosage of sedatives and the monitoring of sedation depth have always been concerned.Therefore,exploring the auxiliary methods that can reduce the dosage of sedatives and the potential indicators that can accurately monitor the sedation depth in realtime are the research directions with great clinical value.Due to the inevitable adverse effects of pharmacologically induced sedation,nonpharmacological treatment,such as acupuncture,has been developed to reduce the dosage of sedatives.Current research has shown that the transcutaneous acupoint electrical stimulation(TAES)combined with sedatives,can reduce the dosage of sedatives and deepen the sedation level.In this thesis,we aimed to quantify the sedative effects of TAES and to investigate the possible brain function mechanisms underly the TAES induced sedation.To achieve this aim,we employed EEG based power spectrum analysis and brain function network analysis.The sedation depth was quantified by the Bispectral Index(BIS),and the acupoints stimulated were: Zusanli(ST36),Shenmen(HT7),and Sanyinjiao(SP6).30-channels of electroencephalograph(EEG)signals were acquired at the resting and TAES stages.First,spectrum analysis was performed on the four frequency bands of delta,theta,alpha and beta(0.5-25 Hz),the permutation entropy index was used to compare the complexity of EEG signals between the two stages.We further calculated four classic graphtheory measures in the delta(0.5-4 Hz)and alpha(9-12 Hz)frequency bands to measure the brain function integration and separation capabilities during these two stages.The results showed that TAES alone can produce a significant sedative effect,the relative power of the delta band in the TAES stage is significantly increased,and that of alpha band is reduced substantially.The permutation entropy index of each channel in the TAES stage has a tendency to decrease,especially for the frontal area,which indicates that the sedative effect induced by TAES is accompanied by a reduction in the complexity of the EEG signal.In addition,brain functional connectivity was significantly reduced during TAES induced sedation.The clustering coefficient,local efficiency,and global efficiency of the alpha band have a tendency to decrease,and the characteristic path length has an increasing trend,but all of them are not significant,and the four measures of the delta are almost similar during these two stages.At present,the sedation depth monitoring system used in the clinic has low time resolution and only collects EEG signals in a part of the brain region,so it cannot accurately reflect the rapid changes of the whole brain consciousness in real time.As the second part of this thesis,we used EEG microstate analysis to distinguish the dynamic changes of propofol induced altered states of consciousness.We collected 60-channel EEG data at three levels of consciousness: awake,moderate sedation,and deep sedation.The depth of sedation was measured by BIS.Microstate templates of three levels of consciousness were obtained by clustering methods,and the parameters that measure the spatiotemporal characteristics of the microstate sequence were calculated.Based on the parameters co-varied with the depth of sedation,a support vector machine(SVM)model was constructed to distinguish different consciousness levels.Finally,the microstates found in the second work were source analyzed by using the Standardized Low-Resolution Electromagnetic Tomography(s LORETA)to identify their cortical generator.The results showed that the optimal number of microstate templates varied with the consciousness levels,and two sedation-specific microstates were found in our research.The SVM classification model was constructed with features showing significantly changed across consciousness levels,and achieved an accuracy of 85%,indicating that the microstate parameters can be utilized for monitoring the sedation depth.The results of source analysis showed that the sedation-specific microstates M8 and M9 have the largest activation area in the frontal regions.The activation area of M4 and M5 did not change with the change of consciousness level.With the deepening of the sedation depth,the brain activation areas corresponding to the rest of the microstates show a back-forward pattern from the occipital region,the parietal region to the cingulate region then to the frontal region.