Many species of birds gradually adapt to urbanization and colonize cities successfully. However, their nest site selection and competitive relationship in an urban community remain little known. Understanding the impact of urbanization on birds and the competitive relationship has important implications for the conservation and management of wildlife in urban ecosystems. Here, we undertook a systematic study to quantify nests in all species of birds in an urbanizing area of Nanchang, China. A total of 363 nests were detected in surveys including 340 nests of 16 bird species and 23 unidentified species nests. We mainly analyzed 5 dominant breeding birds with a sample size of >10 during the two breeding seasons (From April to July in 2016 and 2017), which included the light-vented bulbul, Chinese blackbird, scaly-breasted munia, spotted dove and grey-capped greenfinch. Most birds (93.66%) nested in the tree of artificial green belts, which seems to be the best breeding habitat for urban birds. Our results suggested that birds' breeding success relies on the trade-off between the benefit and the expense of specific stresses from habitats. The nest site selection of birds is also affected by the life habit of urban predators. Furthermore, competition among species can influence their distributions and utilization of environmental resources when birds nest in cities. We confirmed that the niche differentiation of five bird species in an urban environment makes them coexist successfully by utilizing various resources.
Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary care physicians (PCPs), particularly in low-resource settings. Here, to bridge the gaps, we developed an integrated image-language system (DeepDR-LLM), combining a large language model (LLM module) and image-based deep learning (DeepDR-Transformer), to provide individualized diabetes management recommendations to PCPs. In a retrospective evaluation, the LLM module demonstrated comparable performance to PCPs and endocrinology residents when tested in English and outperformed PCPs and had comparable performance to endocrinology residents in Chinese. For identifying referable DR, the average PCP's accuracy was 81.0% unassisted and 92.3% assisted by DeepDR-Transformer. Furthermore, we performed a single-center real-world prospective study, deploying DeepDR-LLM. We compared diabetes management adherence of patients under the unassisted PCP arm (n = 397) with those under the PCP+DeepDR-LLM arm (n = 372). Patients with newly diagnosed diabetes in the PCP+DeepDR-LLM arm showed better self-management behaviors throughout follow-up (P