探讨模糊神经网络在麻醉监测中的应用  

Discuss Application of Fuzzy Neural Network in the Depth of Anesthesia Monitoring

作　　者：李敏 潘慧平 叶志前[1] 

机构地区：[1]浙江大学生物医学工程与仪器科学学院,杭州310027 [2]浙江瑞安人民医院,瑞安325200

出　　处：《中国医疗器械信息》2007年第9期5-10,共6页China Medical Device Information

摘　　要：手术过程中,对患者麻醉深度的监测极其重要,常因不当麻醉而给患者带来心理上的后遗症。因此可靠而无创伤的麻醉深度监测方法是令人期望的。本文探讨了建立模糊神经网络模型以期实现麻醉深度的监测。文中从EEG信号中提取Kc复杂度、近似熵、小波熵三个非线性动力学特征参数,对模糊神经网络进行训练,成功监测病人是否处于麻醉状态,精确度达到97.08%/98.65%(麻醉状态/正常状态)。实验证明,模糊神经网络具有良好的准确性和稳定性,是麻醉深度监测有效而又有潜力的一个工具,具有临床应用价值。During surgery,adequate levels of anesthesia are very important for patient.The inadequate levels of anesthesia can often lead to untoward psychological consequences.So reliable and noninvasive monitoring of the depth of anesthesia is highly desirable.This paper discussed setting up a fuzzy neural network for built and realized to monitor the depth of anesthesia（DOA）.Kc complexity（Kc）,approximate entropy（ApEn）and Wavelet entropy（WE）were extracted from electroencephalogram（EEG）,these parameters were used as input to the fuzzy neural network with one output—DOA.The fuzzy neural network was successfully trained to monitor anesthesia or awake,and the accuracy of output reached 97.08%/98.65%（anesthesia/normal）.It is proved the fuzzy neural network was accurate and robust,was a promising candidate as an effective tool for monitoring of the DOA and would be help to clinical application.

关 键 词：脑电图 麻醉深度 模糊神经网络 信息融合 

分 类 号：R614[医药卫生—麻醉学]