胃电磁信号的检测与分析方法研究

发布时间：2018-04-19 12:01

本文选题：胃磁信号 + 胃电信号　；参考：《东北大学》2014年硕士论文

【摘要】：胃部信号包含胃快波和胃慢波两部分。目前对于胃部信号的检测主要分为两个方向,一个是胃电信号的检测,另一个就是胃磁信号的检测。胃的消化功能是人体的重要机能,胃功能强弱直接影响着人体对所需要营养物质的吸收和利用,与人类的自身健康息息相关。但是至今对这一领域的研究仍然缺乏有效的检测手段,主要是因为胃部系统本身就可认为是一个较为复杂的待检测系统,又由于其自身的独特构造以及所处在人体内的位置,所以胃电磁信号检测领域还需要学者投入更多的精力研究才能得到长足的发展。本文首先进行了胃电信号的检测与分析。胃电采集系统电路的主电路由一级放大电路,带通滤波器与主放大电路组成。并对采集到的胃电信号进行分析与研究。胃磁信号的检测目前主要的依靠手段是超导量子干涉仪(Superconducting Quantum Interference Device, SQUID),但是该仪器体积巨大并且价格昂贵。针对以上的不足之处,本文提出了基于巨磁阻抗效应传感器(Giant Magneto-Impedance sensor, GMI sensor)进行胃磁信号的检测方法,并且对测得的胃磁信号进行胃磁快波信号提取。由于胃磁信号十分微弱并含有噪声,本文中设计了由仪表放大电路与一个同相比例放大电路组成的两级放大电路、滤波电路,有效地滤除了噪音信号。传统的分离方法采用的是盲源分离算法,该方法用于分析胃磁信号时存在精度不高的缺陷,本文给出了小波分析与盲源分离相结合的信号分离方法,对胃磁快慢波进行分离,可有效地分离出快慢波信号。对胃磁信号进行经验模式分解(Empirical Mode Decomposition, EMD),在此基础上进行Hilbert谱与边际谱计算,得到各个频率分量的百分比,以此作为胃动力功能状态分类特征。采用支持向量机(Support Vector Machine, SVM)进行胃磁快波信号的分类,检测胃动力异常。实验结果表明,本文的分类方法具有较高的精度。
[Abstract]:The gastric signal consists of two parts: the fast wave of stomach and the slow wave of stomach.At present, the detection of gastric signals is divided into two directions, one is the detection of gastric electrical signals, the other is the detection of gastric magnetic signals.The digestive function of the stomach is an important function of the human body. The strength of the gastric function directly affects the absorption and utilization of nutrients needed by the human body and is closely related to human health.However, there is still a lack of effective detection methods in this field, mainly because the stomach system itself can be regarded as a more complex system to be detected, and because of its unique structure and position in the human body.Therefore, the field of gastric electromagnetic signal detection still needs scholars to invest more energy in order to get considerable development.In this paper, the detection and analysis of gastric electrical signals are carried out.The main circuit of the gag acquisition system consists of a primary amplifier circuit, a bandpass filter and a main amplifier circuit.The analysis and research of the collected gastric electrical signals were carried out.At present, the main means of detecting gastric magnetic signal is superconducting Quantum Interference device, but this instrument is huge and expensive.In view of the above shortcomings, this paper presents a method for detecting the gastric magnetic signal based on Giant Magneto-Impedance sensor (GMI sensor), and extracts the fast wave signal from the measured signal.Because the gastric magnetic signal is very weak and contains noise, a two-stage amplifier circuit, which is composed of the instrument amplifier circuit and the in-phase proportional amplifier circuit, is designed in this paper. The filter circuit can filter the noise signal effectively.The traditional separation method uses blind source separation algorithm, which has some defects in analyzing gastric magnetic signals. In this paper, a signal separation method combining wavelet analysis and blind source separation is presented to separate fast and slow gastric magnetic signals.The fast and slow wave signals can be separated effectively.On the basis of empirical Mode decomposition of gastric magnetic signal, the Hilbert spectrum and marginal spectrum were calculated, and the percentage of each frequency component was obtained, which was regarded as the classification feature of gastric motility state.The gastric magnetic fast wave signals were classified by support vector machine (SVM) and abnormal gastric motility was detected.The experimental results show that the classification method in this paper has high accuracy.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R57;TN911.23

【相似文献】