Remote sensing imagery is vital for environmental and land use monitoring, necessitating efficient segmentation methods. While deep learning models have become standard for segmentation, they require extensive per-pixel annotations, which are often scarce. High intra-dataset variance also limits the transferability of these models. Although models like SAM offer zero-shot instance-level segmentation, adapting them for semantic segmentation remains challenging. We introduce Text2Seg, a method for remote sensing semantic segmentation that reduces dependency on extensive annotations by automatically generating prompts using visual foundation models (VFMs). This approach enhances generalization across datasets. Experiments on four remote sensing datasets show Text2Seg significantly outperforms the vanilla SAM model, with improvements ranging from 31% to 225%. Our code is available at https://github.com/Douglas2Code/Text2Seg.

Background: Disparities in cardiovascular disease (CVD) outcomes persist across the United States. Social determinants of health play an important role in driving these disparities. The current study aims to identify the most important social determinants associated with CVD mortality over time in US counties. Methods and Results: The authors used the Agency for Healthcare Research and Quality's database on social determinants of health and linked it with CVD mortality data at the county level from 2009 to 2018. The age‐standardized CVD mortality rate was measured as the number of deaths per 100 000 people. Penalized generalized estimating equations were used to select social determinants associated with county‐level CVD mortality. The analytic sample included 3142 counties. The penalized generalized estimating equation identified 17 key social determinants of health including rural–urban status, county's racial composition, income, food, and housing status. Over the 10‐year period, CVD mortality declined at an annual rate of 1.08 (95% CI, 0.74–1.42) deaths per 100 000 people. Rural counties and counties with a higher percentage of Black residents had a consistently higher CVD mortality rate than urban counties and counties with a lower percentage of Black residents. The rural–urban CVD mortality gap did not change significantly over the past decade, whereas the association between the percentage of Black residents and CVD mortality showed a significant diminishing trend over time. Conclusions: County‐level CVD mortality declined from 2009 through 2018. However, rural counties and counties with a higher percentage of Black residents continued to experience higher CVD mortality. Median income, food, and housing status consistently predicted higher CVD mortality.

Geolocating precise locations from images presents a challenging problem in computer vision and information retrieval. Traditional methods typically employ either classification-dividing the Earth's surface into grid cells and classifying images accordingly, or retrieval-identifying locations by matching images with a database of image-location pairs. However, classification-based approaches are limited by the cell size and cannot yield precise predictions, while retrieval-based systems usually suffer from poor search quality and inadequate coverage of the global landscape at varied scale and aggregation levels. To overcome these drawbacks, we present Img2Loc, a novel system that redefines image geolocalization as a text generation task. This is achieved using cutting-edge large multi-modality models (LMMs) like GPT-4V or LLaVA with retrieval augmented generation. Img2Loc first employs CLIP-based representations to generate an image-based coordinate query database. It then uniquely combines query results with images itself, forming elaborate prompts customized for LMMs. When tested on benchmark datasets such as Im2GPS3k and YFCC4k, Img2Loc not only surpasses the performance of previous state-of-the-art models but does so without any model training. A video demonstration of the system can be accessed via this link https://drive.google.com/file/d/16A6A-mc7AyUoKHRH3_WBRToRC13sn7tU/view?usp=sharing

Similar geographic areas often have great variations in population size. In health data management and analysis, it is desirable to obtain regions of comparable population by decomposing areas of large population (to gain more spatial variability) and merging areas of small population (to mask privacy of data). Based on the Peano curve algorithm and modified scale-space clustering, this research proposes a mixed-level regionalization (MLR) method to construct geographic areas with comparable population. The method accounts for spatial connectivity and compactness, attributive homogeneity, and exogenous criteria such as minimum (and approximately equal) population or disease counts. A case study using Louisiana cancer data illustrates the MLR method and its strengths and limitations. A major benefit of the method is that most upper level geographic boundaries can be preserved to increase familiarity of constructed areas. Therefore, the MLR method is more human-oriented and place-based than computer-oriented and space-based.

In December 2019, the coronavirus disease 2019 (COVID-19) pandemic attacked Wuhan, China. The city government soon strictly locked down the city, implemented a hierarchical diagnosis and treatment system, and took a series of unprecedented pharmaceutical and non-pharmaceutical measures. The residents’ access to the medical resources and the consequently potential demand–supply tension may determine effective diagnosis and treatment, for which travel distance and time are key indicators. Using the Application Programming Interface (API) of Baidu Map, we estimated the travel distance and time from communities to the medical facilities capable of treating COVID-19 patients, and we identified the service areas of those facilities as well. The results showed significant differences in service areas and potential loading across medical facilities. The accessibility of medical facilities in the peripheral areas was inferior to those in the central areas; there was spatial inequality of medical resources within and across districts; the amount of community healthcare centers was insufficient; some communities were underserved regarding walking distance; some medical facilities could be potentially overloaded. This study provides reference, in the context of Wuhan, for understanding the spatial aspect of medical resources and residents’ relevant mobility under the emergency regulation, and re-examining the coordination of emergency to improve future planning and utilization of medical facilities at various levels. The approach can facilitate policymakers to assess potential loading of medical facilities, identify low-accessibility areas, and deploy new medical facilities. It also implies that the accessibility analysis can be rapid and relevant even only with open-source data.