基于生理信息熵的人体体成分模型研究

To some extent, human body composition reflects the changes in health status and nutritional status of the human body, and it has important significance for disease prevention.Currently, the building of human body composition model is mainly based on various body composition prediction models based on bioelectrical impedance measurement method(BIA).It can only express segment impedance, but it is difficult to reflect the trunk of the body segments composition variation . The project intends to combine the information entropy theory with results of the human body composition models. Through studying eight bioelectrical impedance model based on trunk segment and various physiological parameters of the criteria, we can build Physiological model of the human body composition based on information entropy, and then study the human body composition solution strategy model based on physiological information entropy. Finally, design the human body composition model simulations based on physiological information entropy for data comparison analysis, verification and analysis of results. Human body composition model method based on information entropy effectively overcome some shortcomings , such as, the presence of more than the number of parameters, severe affiliation between each parameter, to meet the needs of the human body composition prediction.

人体体成分的变化在一定程度上反映了身体健康状况的改变，对人体营养状况及疾病预防有着重要意义。鉴于目前人体体成分模型的构建主要建立在各种基于生物电阻抗测量方法（BIA）的经验体成分预测模型的基础上，只能表达人体分段阻抗的生理特征，难以反映躯干细分的体成分变化规律，本项目拟将信息熵理论融入已有人体体成分建模的研究成果，通过研究基于躯干细分的八段生物电阻抗模型求解，以及各种生理特征参数的信息熵准则，构建基于生理信息熵的人体体成分模型。进而研究基于生理信息熵的人体体成分模型的求解策略。最后，设计基于生理信息熵的人体体成分模型仿真实验，进行数据对比分析，验证及结果分析。基于生理信息熵的人体体成分模型方法有效克服现有人体体成分存在参数数量多、各参数之间关联严重等缺点，满足了人体体成分预测的需要。

人体体成分的变化在一定程度上反映了身体健康状况的改变，对人体营养状况及疾病预防有着重要意义。鉴于目前人体体成分模型的构建主要建立在各种基于生物电阻抗测量方法（BIA）的经验体成分预测模型的基础上，只能表达人体分段阻抗的生理特征，难以反映躯干细分的体成分变化规律，本项目拟将信息熵理论融入已有人体体成分建模的研究成果，通过研究基于躯干细分的八段生物电阻抗模型求解，以及各种生理特征参数的信息熵准则，构建基于生理信息熵的人体体成分模型。进而研究基于生理信息熵的人体体成分模型的求解策略。最后，设计基于生理信息熵的人体体成分模型仿真实验，进行数据对比分析，验证及结果分析。基于生理信息熵的人体体成分模型方法有效克服现有人体体成分存在参数数量多、各参数之间关联严重等缺点，满足了人体体成分预测的需要。