癫痫脑电移动监测系统设计

发布时间：2018-08-14 19:31

【摘要】：癫痫是一种严重的中枢神经系统紊乱疾病,以发作的反复性和不可预知性为特征。脑电(EEG)信号是监测和诊断癫痫发作的一个重要途径。然而,传统EEG监测只能在医院完成,不利于癫痫患者的日常监测,又由于EEG信号的复杂性,给人工诊断癫痫发作带来很大挑战。本文提出了一套综合性的移动医疗监测系统,对EEG信号进行便携式采集、传输,并根据EEG信号的信息对癫痫发作进行自动分析和检测。系统包括两个主要模块:前端模块和后端模块。前端模块是一套基于ADS1299芯片的便携式EEG信号采集前端,包括四个模块:信号采集、控制、数据存储和数据传输。摒弃了以往脑电采集设备所必备的模拟和数字两大电路部分的设计方式,而是采用了目前比较先进的ADS1299芯片对EEG进行放大、滤波、采样等其他多种操作。该套采集设备没有采用高频高性能的控制芯片,而是采用将数据处理程序放到上位机的方式,这样只需要采用更低处理频率的MCU,既保持着应有的功能,降低了整套系统价格,也全面发挥着后端模块的优势。与后端模块的连接方面,考虑到设备的便携式特点,该设备摒弃了传统的USB有线连接方式,采用WIFI方式可以方便的与多种类的上位机进行自由的连接,扩展了有效连接距离。在电源方面,该设备采用目前流行的移动电源,有效消除50Hz工频干扰,进一步增强设备的稳定性。后端模块包括预处理、特征提取和分类。本文以公共数据库中相关癫痫EEG数据为例,首先对原始EEG信号进行带通和陷波滤波,去除伪迹和工频干扰;然后进行小波分解,进而根据小波能量熵进行特征提取,最后采用支持向量机将信号分为未发作和发作两种状态,完成癫痫发作的探测。仿真结果证明了所提出方案的有效性。
[Abstract]:Epilepsy is a severe disorder of the central nervous system characterized by recurrent and unpredictable seizures. EEG (EEG) signal is an important way to monitor and diagnose epileptic seizures. However, traditional EEG monitoring can only be done in hospitals, which is not conducive to the routine monitoring of epilepsy patients, and because of the complexity of EEG signals, it brings a great challenge to the artificial diagnosis of epileptic seizures. This paper presents a comprehensive mobile medical monitoring system, which can collect and transmit EEG signals in a portable manner, and analyze and detect epileptic seizures automatically according to the information of EEG signals. The system includes two main modules: front-end module and back-end module. The front-end module is a portable EEG signal acquisition front-end based on ADS1299 chip, which includes four modules: signal acquisition, control, data storage and data transmission. The design method of analog and digital circuits necessary for EEG acquisition equipment is abandoned, and the more advanced ADS1299 chip is used to amplify, filter and sample EEG. Instead of using the high frequency and high performance control chip, the acquisition equipment adopts the method of putting the data processing program into the upper computer, which only needs MCU with lower processing frequency, which not only keeps the proper function, but also reduces the price of the whole system. Also give full play to the advantages of the back-end module. In connection with the back-end module, considering the portable characteristic of the device, the device abandons the traditional USB wired connection mode, and adopts the WIFI mode to connect freely with many kinds of host computer conveniently, and extends the effective connection distance. In the aspect of power supply, the equipment adopts the current popular mobile power supply, effectively eliminates the 50Hz power frequency interference, and further enhances the stability of the equipment. The back-end module includes preprocessing, feature extraction and classification. In this paper, we take the relevant epileptic EEG data in the common database as an example. First, the original EEG signal is filtered by bandpass and notch wave to remove artifact and power frequency interference, then wavelet decomposition is carried out, and then the feature extraction is carried out according to wavelet energy entropy. Finally, support vector machine (SVM) is used to detect epileptic seizures. Simulation results show the effectiveness of the proposed scheme.
【学位授予单位】：天津职业技术师范大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R742.1

【相似文献】