生物组织光学参数与热物性参数的相关性研究

基于组织光学和人工神经网络理论,提出了一种全新的确定生物体热物性参数的方法.利用积分球系统测量介质组织光学特性参数,以该特性参数作为人工神经网络输入变量,以热物性参数为目标变量,利用神经网络的自学习系统,最终建立光学特性参数与热物性参数间的非线性映射一黑箱模型.为验证模型,测量了多种热物性参数已知物质的光学特性参数,然后由黑箱模型得到热物性参数计算值,计算值和已知真实值的相对误差均在5%以内.该方法可用于测量生物体的各种热物性参数.

RESEARCH ON THE CORRELATION BETWEEN THERMOPHYSICAL PARAMETERS AND OPTICAL PARAMETERS OF ORGANISM

Based on the theories of tissue optics and artificial neural network,a novel method of mea- suring thermophysical parameters was brought forward.A nonlinear mapping-black-box model-was established by self-study of the artificial neural network whose input variable were optical parameters which were measured by the system of double integrating spheres and whose target variable were ther- mophysical parameters.This model had been used to calculate the thermophysical parameters of some materials based on their measured optical variables.Comparison between the calculated and accessible thermophysical parameters from literatures,the maximum relative error was 5%.In principle,this model not only could be used to measure these thermophysical parameters,but also developed a new means for the measurement of parameters in other domains.