基于近红外光谱的人体血液酒精含量无创检测方法研究

研发一种人体血液酒精含量无创检测方法迫在眉睫.近些年来,随着近红外光谱在定量分析化学物质含量方面的迅速发展,以及人工神经网络在电子信息领域的延伸,可以利用近红外光谱的方法,在MATLAB软件环境下,用神经网络方法进行建模,然后将适当的样本数据训练后,最终预测人体血液酒精含量.就如何在BP神经网络和RBF神经网络建模的基础上,对人体血液酒精含量的数据处理作了详细论述.

Noninvasive test of human blood alcohol content based on the near infrared spectra

Developing a noninvasive test method about human blood alcohol content is around the comer. In recent years, with the quick development of the near infrared spectra in the quantitative analysis of the chemical content and the outspread of artificial neural network in the field of electronic information, we can use the near infrared spectrum research methods in the MATLAB software environment to build model with neural network method. After training the appropriate sample data, we can predict human blood alcohol concentration. Based on the BP neural network and RBF neural network model, this paper discussed the data treatment of human body blood alcohol concentration in detail.