【摘要】 应用支持向量机(SVM)对生物电阻抗测定人体内脏脂肪过程进行测量建模,解决信号复杂,受影响因素多,难以建立精确预测模型问题。为提高预测精度,引入人体腹部形状作为测量人体内脏脂肪的相关特征参数,将SVM参数的选择和输入变量的选取看作组合优化问题,通过AIC信息准则构造组合目标优化函数,采用粒子群算法进行目标函数搜索,提高了搜索效率。通过仿真研究表明,所提基于SVM的生物电阻抗人体内脏脂肪含量测量方法具有良好的性能。更多还原

【Abstract】 Due to complex signals and many interfering factors in the process of assessing the viscera fat using bio-electrical impedance,it is hard to get an accurate predicting model.The selection and simplification of the feature pa-rameters about the human viscera fat is discussed.The viscera shape is adopted as one of the important feature parameters of viscera fat.A compound optimal objective function based on Akaike information criterion is constructed.The PSO optimal algorithm is used to search the optimal value of the objective function to improve the efficiency.The result of the simulation shows the method based on SVM has good capability to measure the human viscera fat.更多还原