School of Information Technology, Deakin University, Geelong, VIC, Australia.
Guangzhou Xinhua University, Dongguan, China.
Front Public Health. 2024 Jun 5;12:1357715. doi: 10.3389/fpubh.2024.1357715. eCollection 2024.
To enhance the precision of evaluating the impact of urban environments on resident health, this study introduces a novel fuzzy intelligent computing model designed to address health risk concerns using multi-media environmental monitoring data.
Three cities were selected for the study: Beijing ( City), Kunming ( City), and Wuxi ( City), representing high, low, and moderate pollution levels, respectively. The study employs a Fuzzy Inference System (FIS) as the chosen fuzzy intelligent computing model, synthesizing multi-media environmental monitoring data for the purpose of urban health risk assessment.
(1) The model reliably estimates health risks across diverse cities and environmental conditions. (2) There is a positive correlation between PM2.5 concentrations and health risks, though the impact of noise levels varies by city. In cities , , and , the respective correlation coefficients are 0.65, 0.55, and 0.7. (3) The Root Mean Square Error (RMSE) values for cities , , and , are 0.0132, 0.0125, and 0.0118, respectively, indicating that the model has high accuracy. The R values for the three cities are 0.8963, 0.9127, and 0.9254, respectively, demonstrating the model's high explanatory power. The residual values for the three cities are 0.0087, 0.0075, and 0.0069, respectively, indicating small residuals and demonstrating robustness and adaptability. (4) The model's p-values for the Indoor Air Quality Index (IAQI), Thermal Comfort Index (TCI), and Noise Pollution Index (NPI) all satisfy < 0.05 for the three cities, affirming the model's credibility in estimating health risks under varied urban environments.
These results showcase the model's ability to adapt to diverse geographical conditions and aid in the accurate assessment of existing risks in urban settings. This study significantly advances environmental health risk assessment by integrating multidimensional data, enhancing the formulation of comprehensive environmental protection and health management strategies, and providing scientific support for sustainable urban planning.
简介：为了提高评估城市环境对居民健康影响的准确性，本研究引入了一种新的模糊智能计算模型，该模型旨在使用多媒体环境监测数据来解决健康风险问题。
方法：选择了三个城市进行研究：北京（城市）、昆明（城市）和无锡（城市），分别代表高、低和中污染水平。该研究采用模糊推理系统（FIS）作为所选的模糊智能计算模型，综合多媒体环境监测数据进行城市健康风险评估。
结果：（1）该模型能够可靠地估计不同城市和环境条件下的健康风险。（2）PM2.5 浓度与健康风险之间存在正相关关系，但噪声水平的影响因城市而异。在城市、和中，相应的相关系数分别为 0.65、0.55 和 0.7。（3）城市、和的均方根误差（RMSE）值分别为 0.0132、0.0125 和 0.0118，表明该模型具有较高的准确性。三个城市的 R 值分别为 0.8963、0.9127 和 0.9254，表明该模型具有较高的解释能力。三个城市的残差值分别为 0.0087、0.0075 和 0.0069，表明残差较小，表明该模型具有稳健性和适应性。（4）模型对室内空气质量指数（IAQI）、热舒适指数（TCI）和噪声污染指数（NPI）的 P 值在三个城市均满足 <0.05，证实了该模型在不同城市环境下评估健康风险的可信度。
讨论：这些结果展示了该模型适应不同地理条件的能力，并有助于准确评估城市环境中的现有风险。本研究通过整合多维数据，提高了环境健康风险评估的能力，制定了更全面的环境保护和健康管理策略，并为可持续城市规划提供了科学支持。
