Advanced signal processing and control in anaesthesia.

Advanced signal processing and control in anaesthesia.
复制标题

麻醉中的先进信号处理和控制。

DOI:
--
复制
发表时间:
2002
期刊:
影响因子:
--
通讯作者:
C. Sofia
C. Sofia
中科院分区:
--
文献类型:
--
作者:
D. Nunes;C. Sofia

文献摘要

被引文献

相似文献

本论文包括三个主要阶段:麻醉深度(DOA)的分类;在外科手术过程中模拟典型患者的行为;以及同时使用丙泊酚和瑞芬太尼控制DOA。在手术室收集的临床数据被用于本项目。 采用多分辨率小波分析对听觉诱发电位(AEP)信号进行特征提取。这些功能被分为不同的DOA水平,使用模糊关系分类器(FRC)。FRC使用模糊聚类和模糊关系组合。FRC具有良好的性能,并且能够区分DOA水平。 一个混合的病人模型被开发用于麻醉的诱导和维持阶段。采用基于自适应网络的模糊推理系统对TSK模糊模型进行自适应调整,该模型将收缩压(SAP)、心率(HR)和小波提取的AEP特征与丙泊酚和瑞芬太尼的有效浓度相关联。手术刺激对SAP和HR的影响,以及瑞芬太尼的镇痛特性由Mamdani模糊模型描述,与麻醉师合作构建。该模型被证明是适当的,反映了药物和手术刺激的影响。 设计了一种用于丙泊酚和瑞芬太尼同时给药的多变量模糊控制器。该控制器是基于语言的规则,相互作用的三个决策表,其中之一是一个模糊PI控制器。根据DOA水平和手术刺激确定两种药物的输注速率。瑞芬太尼根据所需的镇痛水平及其与丙泊酚的协同作用进行滴定。该控制器能够在不同的条件下充分达到并保持目标DOA水平。 总体而言,可以对异丙酚和瑞芬太尼之间的相互作用进行建模,并成功使用该模型开发麻醉闭环系统。
This thesis comprises three major stages: classification of depth of anaesthesia (DOA); modelling a typical patient’s behaviour during a surgical procedure; and control of DOAwith simultaneous administration of propofol and remifentanil. Clinical data gathered in theoperating theatre was used in this project. Multiresolution wavelet analysis was used to extract meaningful features from the auditory evoked potentials (AEP). These features were classified into different DOA levels using a fuzzy relational classifier (FRC). The FRC uses fuzzy clustering and fuzzy relational composition. The FRC had a good performance and was able to distinguish between the DOA levels. A hybrid patient model was developed for the induction and maintenance phase of anaesthesia. An adaptive network-based fuzzy inference system was used to adapt Takagi-Sugeno-Kang (TSK) fuzzy models relating systolic arterial pressure (SAP), heart rate (HR), and the wavelet extracted AEP features with the effect concentrations of propofol and remifentanil. The effect of surgical stimuli on SAP and HR, and the analgesic properties of remifentanil were described by Mamdani fuzzy models, constructed with anaesthetist cooperation. The model proved to be adequate, reflecting the effect of drugs and surgical stimuli. A multivariable fuzzy controller was developed for the simultaneous administration of propofol and remifentanil. The controller is based on linguistic rules that interact with three decision tables, one of which represents a fuzzy PI controller. The infusion rates of the two drugs are determined according to the DOA level and surgical stimulus. Remifentanil is titrated according to the required analgesia level and its synergistic interaction with propofol. The controller was able to adequately achieve and maintain the target DOA level, under different conditions. Overall, it was possible to model the interaction between propofol and remifentanil, and to successfully use this model to develop a closed-loop system in anaesthesia.