Combined use of principal component analysis and artificial neural network approach to improve estimates of PM2.5 personal exposure: A case study on older adults
Combined use of principal component analysis and artificial neural network approach to improve estimates of PM2.5 personal exposure: A case study on older adults
复制标题
结合使用主成分分析和人工神经网络方法来改进 PM2.5 个人暴露估计：针对老年人的案例研究
DOI：
10.1016/j.scitotenv.2020.138533
复制
发表时间：
2020
影响因子：
9.8
通讯作者：
Hai Yu
中科院分区：
文献类型：
--
作者：
Shuang Gao;Hong Zhao;Zhipeng Bai;Bin Han;Jia Xu;Ruojie Zhao;Nan Zhang;Li Chen;Xiang Lei;Wendong Shi;Liwen Zhang;Penghui Li;Hai Yu
由于空气污染物PM2.5对人体的呼吸系统和心血管系统有不良影响，因此在流行病学研究中需要对PM2.5的暴露量进行准确的估计，以评估其对健康的影响。然而，传统的个人采样是耗时且成本高的。因此，需要建模技术来准确预测个人对PM2.5的暴露水平。本研究在中国北方空气污染严重的天津市，对117名60岁以上的老年人进行了室内、室外和个人PM2. 5采样。人工神经网络（ANN）模拟18个变量，可能会增加老年人的暴露水平。四个建模技术，包括时间积分活动建模，蒙特卡罗模拟，人工神经网络建模，并结合使用主成分分析（PCA）和人工神经网络模型，被用来评估他们的能力预测真实的暴露值的PM2. 5。传统的时间加权活动模型与实测值的相关性最低，冬季和夏季的R2分别为0.57和0.42。对于Monte Carlo模拟，预测值和真实的暴露值之间的相关性很高（冬季和夏季的R2分别为0.93和0.92）。与简单的人工神经网络模型相比，主成分分析和人工神经网络的结合使用产生了最准确的结果与R2为0.99和RMSE低于15。由于PCA-ANN模型的输入变量信息可以从调查问卷和固定的空气质量监测点获得，这种技术在预测个人暴露水平的空气污染物，因为不需要额外的浓度测量显示了很大的潜力。
Accurate exposure estimate of the air pollutant PM2.5is required to evaluate its health impacts in epidemiological studies, due to its adverse effects on human's respiratory and cardiovascular systems. However, traditional personal sampling is time and cost consuming. Thus, modeling techniques are needed to accurately predict the personal exposure level to PM2.5. In this study, a total of 117 older adults over 60 were recruited in Tianjin, a heavily polluted city in northern China, for indoor, outdoor and personal PM2.5sampling. Eighteen variables which may increase the exposure level of older adults were recorded for artificial neural network (ANN) simulation. Four modeling techniques, including time-integrated activity modeling, Monte Carlo simulation, ANN modeling, and combined use of principal component analysis (PCA) and ANN model, were used to evaluate their ability for predicting real exposure values of PM2.5. The results of traditional time-weighted activity modeling showed the lowest correlation with measured values with R2of 0.57 and 0.42 in winter and summer, respectively. For Monte Carlo simulation, high correlation was obtained (R2of 0.93 and 0.92 in winter and summer, respectively) between percentiles of the predicted and the real exposure values. Compared with the simple ANN models, the combined use of PCA and ANN produced the most accurate results with R2of 0.99 and RMSE lower than 15. Since the information of the input variables for the PCA-ANN model can be obtained from the questionnaire and fixed air quality monitoring sites, this technique shows a great potential in predicting personal exposure level to the air pollutant because no additional concentration measurement is needed.