论文部分内容阅读
将现代科学技术与传统中医理论相结合,提出一种基于小波神经网络技术的人体经穴电位信号分析方法。目的是利用神经网络的分类能力区分和辨别中风病人与正常人不同身体生理状态。首先,采集人体手部6个原穴(神门,太渊,大陵,合谷,阳池,腕骨)处的电位信号,再将得到的信号进行小波去噪及小波多分辨率分析,进而选择性地提取出某些小波系数,用以计算相应的能量系数、高低频能量系数之比和各层能量系数方差。计算结果经归一化之后,作为特征向量输入到以morlet小波作为隐函数的小波神经网络,进行训练和测试。结果显示神经网络表现满足预期要求,说明本方法区分和辨别中风病人与正常人不同身体生理状态的可行性、有效性、高效性。
Combining modern science and technology with traditional Chinese medicine theory, a method of human acupuncture point signal analysis based on wavelet neural network technology is proposed. The purpose is to use neural network classification ability to distinguish and identify stroke patients and normal different physical state. First of all, we collected the potential signals of 6 original holes (Shenmen, Taiyuan, DaLing, Hegu, Yangchi, and carpal bones) of the human hand, and then selected the signals by wavelet denoising and wavelet multiresolution analysis Some wavelet coefficients are extracted qualitatively to calculate the corresponding energy coefficient, the ratio of high and low frequency energy coefficients, and the variance of energy coefficients of each layer. After normalization, the result is input to the wavelet neural network with Morlet wavelet as an implicit function as the feature vector for training and testing. The results show that the neural network performance meets the expected requirements, indicating the feasibility, effectiveness and efficiency of this method in distinguishing and identifying different physical and physiological states of stroke patients and normal subjects.