High-resolution remotely sensed imageries have been widely employed to detect urban villages (UVs) in highly urbanized regions, especially in developing countries. However, the understanding of the potential impacts of spatially and temporally differentiated urban internal development on UV detection is still limited. In this study, a partition-strategy-based framework integrating the random forest (RF) model, object-based image analysis (OBIA) method, and high-resolution remote sensing images was proposed for the UV-detection model. In the core regions of Guangzhou, four original districts were re-divided into five new zones for the subsequent object-based RF-detection of UVs with a series features, according to the different proportion of construction lands. The results show that the proposed framework has a good performance on UV detection with an average overall accuracy of 90.23% and a kappa coefficient of 0.8. It also shows the possibility of transferring samples and models into a similar area. In summary, the partition strategy is a potential solution for the improvement of the UV-detection accuracy through high-resolution remote sensing images in Guangzhou. We suggest that the spatiotemporal process of urban construction land expansion should be comprehensively understood so as to ensure an efficient UV-detection in highly urbanized regions. This study can provide some meaningful clues for city managers identifying the UVs efficiently before devising and implementing their urban planning in the future.

Increasingly stricter and wider official efforts have been made by multilevel Chinese governments for seeking the improvements of the environment and public health status. However, the contributions of these efforts to environmental changes and spatiotemporal variations in some environmental diseases have been seldom explored and evaluated. Gastric cancer mortality (GCM) data in two periods (I: 2004–2006 and II: 2012–2015) was collected for the analysis of its spatiotemporal variations on the grid scale across S County in Central China. Some environmental and socioeconomic factors, including river, farmlands, topographic condition, population density, and gross domestic products (GDP) were obtained for the exploration of their changes and their relationships with GCM’s spatiotemporal variations through a powerful tool (GeoDetector, GD). During 2004–2015, S County achieved environmental improvement and socioeconomic development, as well as a clear decline of the age-standardized mortality rate of gastric cancer from 35.66/10⁵ to 23.44/10⁵. Moreover, the GCM spatial patterns changed on the grid scale, which was spatially associated with the selected influencing factors. Due to the improvement of rivers’ water quality, the distance from rivers posed relatively larger but reversed impacts on the gridded GCM. In addition, higher population density and higher economic level (GDP) acted as important protective factors, whereas the percentage of farmlands tended to have adverse effects on the gridded GCM in period II. It can be concluded that the decline of GCM in S County was spatiotemporally associated with increasingly strengthened environmental managements and socioeconomic developments over the past decade. Additionally, we suggest that more attentions should be paid to the potential pollution caused by excessive pesticides and fertilizers on the farmlands in S County. This study provided a useful clue for local authorities adopting more targeted measures to improve environment and public health in the regions similar to S County.

Epidemiological features of esophageal cancer (EC), as well as their associations with potential influencing factors in a city, have seldom been seldom explored on a fine scale. The EC death cases in Guangzhou city during 2012−2017 were collected to describe the epidemiological characteristics such as EC mortality rate (ECMR) and health-seeking behaviors of deaths. Potential influencing factors, including socioeconomic conditions (population density, gross domestic product density), medical resources, and ageing degree were also gathered for exploring their relationships with the epidemiological characteristics of EC. A total of 2,409 EC deaths were reported during 2012−2017 in Guangzhou with an age-standardized ECMR of 3.18/10⁵. The prevalence of EC in Guangzhou was spatially featured and was divided into three regions with obvious differentiated ECMR (ECMR of 6.41/10⁵ in region A, ECMR of 5.51/10⁵ in region B, ECMR of 2.56/10⁵ in region C). The street/town-level ECMR was spatially clustered in Guangzhou city, especially two clusters of streets/towns with high ECMR were highlighted in region A and B respectively. Meanwhile, demographic features including gender gap, death age, temporal interval between diagnosis and death, health-seeking behaviors were remarkably different among the three regions. Moreover, health-seeking behaviors (e.g., the proportion of hospital deaths) of the EC deaths were obviously influenced by medical institution occupancy rate and socioeconomic conditions at street/town level. In addition, the street/town-level ECMR was significantly associated with ageing degree across Guangzhou city (r = 0.466, p < 0.01), especially in region A (r = 0.565, p < 0.01). In contrast, the ECMR in region B was closely related to population density (r = −0.524, p < 0.01) and gross domestic product density (r = −0.511, p < 0.01) when the ageing degree was controlled, while these associations were weak in region C. The epidemiological characteristics of EC in Guangzhou city were spatially featured and potentially associated with socioeconomic conditions, medical resources and ageing degree on a fine scale across Guangzhou city. This study could provide scientific basis for local authorities to implement more targeted EC interventions.