Zike Hu’s research while affiliated with Jiangxi Normal University and other places

In this study, an innovative prediction model based on the integration of variational modal decomposition (VMD), particle swarm optimization (PSO) algorithm and bidirectional long and short-term memory network (BiLSTM) is proposed to address the mechanism of economic and social factors on the health life expectancy of the population and prediction problems. The adaptive modal decomposition of complex time-series features by VMD algorithm, combined with the global optimization of key parameters of BiLSTM network by PSO algorithm, effectively solves the limitations of the traditional model in nonlinear data feature extraction and long-term dependency capturing. The empirical results show that the model achieves MAE values of 2.7335 and 2.8195 in the training and test sets, respectively, and the error in the test set is only slightly increased by 0.86 compared with the training set, while the R index in the test set is improved by 0.03 compared with the training set, which shows that the model maintains high prediction accuracy with good generalization ability. The visualization analysis further confirms that the model predictions are highly consistent with the actual healthy life expectancy data in terms of spatial distribution, especially in the fluctuating intervals of the key economic and social indicators, which demonstrates a robust prediction performance. This study not only provides a new quantitative analysis framework for the influence mechanism of healthy life expectancy, but also constructs a prediction model with important application value in the simulation of the effect of public health policies and optimization of health resource allocation, etc. The research results can provide scientific basis for the government to formulate healthy aging policies and promote the development of healthy and equitable development, which is of positive practical guidance significance for the realization of the strategic goal of Healthy China 2030.

Regional haze pollution is a serious atmospheric environmental problem in China, it seriously affects residents’ health. This paper uses ArcGIS software to analyse the spatial and temporal distribution maps of haze pollution and residents’ health in China, and empirically examines the spatial non-linear effects of haze pollution on residents’ health by using spatial econometric models and threshold models. The results are shown as follows: First, the level of haze pollution in China is gradually decreasing, and the level of residents’ health is gradually improving. It is worth noting that the areas with high level of haze pollution are mainly located on the southeast side of the Hu line. Coincidentally, the areas with high level of residents’ health are also mainly located in the southeast side of the Hu line. Second, haze pollution has significant spatial spillover effect on residents’ health, with inter-regional spillover effect being much greater than intra-regional spillover effect. Third, economic development increases the health risk from haze pollution, while environmental regulation does not reduce the health risk from haze pollution. The effect of population agglomeration on the health risk from haze pollution shows a trend of improvement followed by deterioration, while the effect of industrial agglomeration on the health risk from haze pollution shows a trend of deterioration followed by improvement. Therefore, this paper provides valuable insights for the ongoing and comprehensive efforts to fight for a blue sky and promote the implementation of the national strategies of “Beautiful China” and “Healthy China”.