No abstract

No abstract

It is widely accepted that some air pollutants are related to lung cancer prevalence. An effective method is proposed to quantitatively evaluate the effects of air pollutants and the interactions between them. The method consisted of three parts: data decomposition, comparable data generation and relationship inference. Firstly, very limited monitoring data published by Geographic Information System were applied to calculate the inhalable air pollution of relatively massive patient samples. Then the investigated area was partitioned into a number of districts, and the comparable data containing air pollutant concentrations and lung cancer prevalence in all districts were generated. Finally, the relationships between pollutants and lung cancer prevalence were concluded by an information fusion tool: Choquet integral. As an example, the proposed method was applied in the investigation of air pollution in Tianjin, China. Overall, SO, O and PM were the top three factors for lung cancer. And there was obvious positive interaction between O and PM and negative interaction among SO, O and PM. The effect of SO on men was larger than on women. O and SO were the most important factors for the adenocarcinoma and squamous cell carcinoma, respectively. The effect of SO or NO on squamous cell carcinoma is obviously larger than that on adenocarcinoma, while the effect of O or PM on adenocarcinoma is obviously larger than that on squamous cell carcinoma. The results provide important suggestions for management of pollutants and improvement of environmental quality. The proposed method without any parameter is general and easily realized, and it sets the foundation for further researches in other cities/countries. Implications: For total lung cancer prevalence, male and female lung cancer prevalence, and adenocarcinoma and squamous cell carcinoma prevalence, the proposed method not only quantify the effect of single pollutant (SO, NO, CO, O, PM, and PM) but also reveals the correlations between different pollutants such as positive interaction or negative interaction. The proposed method without any geographic predictor and parameter is much easier to realize, and it sets the foundation for further research in other cities/countries. The study results provide important suggestions for the targeted management of different pollutants and the improvement of human lung health.

Rain is a common weather phenomenon that severely degrades outdoor image quality, affecting information extraction. The existing methods can be roughly classified as model-based and data-driven approaches. Model-based methods tend to utilize complex but limiting priors, resulting in optimization difficulty. Data-driven techniques emphasize the establishment of a network architecture and strongly depend on training pairs; hence, they become invalid in practical scenarios. To mitigate these problems, we develop a flexible collaborative layer projection framework for efficient and effective singleimage rain streak removal task. We introduce a solution space to standardize fidelity without loss of generality to build a general deraining model. Then, a Collaborative Layer Projection Framework (CLPF) is presented for solving this model. Using the projection framework, various types of techniques (e.g., learnable architectures and optimization models) can be easily integrated to realize the desired performance. Extensive evaluations of synthetic and real rain images demonstrate that the proposed method outperforms state-of-the-art methods. In addition, we extend the method to other visual areas, such as video deraining and image dehazing. Our framework also performs relatively well on these issues.

No abstract