静息态脑电在意识诊断中的临床应用

Clinical Application of Resting EEG in Consciousness Diagnosis

意识障碍患者的精确判定对制定诊疗计划和预测恢复结果具有重要意义,研究可靠的意识检测方法十分必要。50名意识障碍患者(25名植物状态(VS)患者和25名微意识状态(MCS)患者)被纳入研究,利用脑电双频指数(BIS)监护仪采集脑电数据,并计算信号复杂度和相对功率;通过双样本t检验分析MCS和VS组间脑电特征差异,利用皮尔森相关系数分析脑电特征与临床评分的量化关系,探索区分不同意识状态的脑电特征。利用以上脑电特征构建基于决策树的意识分类模型,应用于意识障碍患者的临床辅助诊断。排序熵(PE)、排序Lempel-Ziv复杂度(PLZC)、gamma频段相对功率在VS和MCS间存在显著差异(PE:0.71±0.07 vs 0.75±0.07,P<0.01;PLZC:0.53±0.07 vs 0.56±0.06,P<0.01;gamma:0.13±0.07 vs 0.16±0.06,P<0.01),且PE与临床评分相关性最高(r=0.81,P<0.001)。基于PE构建的意识分类模型的ROC曲线下的面积(AUC)和准确度(ACC)(AUC=0.931,ACC=0.92)优于BIS构建的分类模型(AUC=0.905,ACC=0.90)的相应参数,表明静息态脑电有望成为意识诊断的重要工具。

The accurate diagnosis of patients with disorder of consciousness(DOC)is of great significance for the treatment plan and outcome,so it is very necessary to develop a reliable method to assess the level of consciousness.In this work,fifty patients diagnosed with DOC(25 vegetable state(VS)patients and 25 minimally consciousness state(MCS)patients)were enrolled.The EEG data were obtained through Bispectral index(BIS)monitor;information complexity and relative power were calculated.The differences of EEG characteristics between MCS and VS groups were analyzed by two sample t-test,and the quantitative relationship between EEG characteristics and clinical scores was analyzed by Pearson correlation analysis.Exploring EEG characteristics to distinguish the states of consciousness.The EEG characteristics were used to build machine-learning model and explore its potential in clinical diagnosis.The results showed that permutation entropy(PE),permutation Lempel-Ziv complexity(PLZC)and the relative power of gamma band were able to distinguish different states of consciousness(PE:0.71±0.07,0.75±0.07,P<0.01;PLZC:0.53±0.07,0.56±0.06,P<0.01;gamma:0.13±0.07,0.16±0.06,P<0.01).PE shows the highest correlation(r=0.81,P<0.05).The area under the ROC curve(AUC)and accuracy(ACC)of consciousness classification model based on PE(AUC=0.931,ACC=0.92)was better than that of BIS(AUC=0.905,ACC=0.90).In conclusion,the resting EEG can be used as an important method for the diagnosis of consciousness.