基于BP神经网络的最大摄氧量预测方法研究

Research on maximum oxygen uptake prediction based on BP neural network

心肺功能能力是影响人体健康的重要指标之一,其中心肺运动试验中所测得的最大摄氧量是评价心肺功能的金标准。然而心肺运动试验所需的设备昂贵,方法复杂,测试场地具有一定的局限性。提出使用BP神经网络来建立预测模型,结合6 min上下楼梯试验中的主要指标来预测人体最大摄氧量。首先验证6 min上下楼梯试验中上下楼梯台阶数与最大摄氧量之间的相关性;之后讨论其它生理指标与最大摄氧量之间的相关性,用以确定BP神经网络预测模型输入量的选择;最后使用实验数据对网络进行训练,通过预测结果分析调整网络参数,从而得到最优的网络模型。MATLAB仿真分析结果证实了将BP神经网络用于人体最大摄氧量估计的可行性和有效性。

Cardiopulmonary function is an important index that affects human health.The maximum oxygen intake measured in cardiopulmonary exercise test is the gold standard for evaluating cardiopulmonary function.However,the equipment required for cardiopulmonary exercise test is expensive and the method is complex.The test site has certain limitations.Therefore,BP neural network is used to establish the prediction model,and the main index of 6 min up and down stairs test is used to predict the maximum oxygen intake of the human body.Firstly,the correlation between the steps number of stairs up and down and the maximum oxygen intake was verified.Then the correlation between other physiological indexes and the maximum oxygen intake is discussed to determine the input quantity of BP neural network prediction model.Finally,the network is trained by using experimental data,and the network parameters are adjusted by predictive analysis,so the optimal network model is obtained.The results of MATLAB simulation confirm the feasibility and effectiveness of BP neural network in estimating maximum oxygen intake in human body.