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

With over 350,000 cases occurring each year, out-of-hospital cardiac arrest (OHCA) remains a severe public health concern in the United States. The correct and timely use of automated external defibrillators (AEDs) has been widely acknowledged as an effective measure to improve the survival rate of OHCA. While general guidelines have been provided by the American Heart Association (AHA) for AED deployment, the lack of detailed instructions hindered the adoption of such guidelines under dynamic scenarios with various time and space distributions. Formulating the AED deployment as a location optimization problem under budget and resource constraints, we proposed an overlayed spatio-temporal optimization (OSTO) method, which accounted for the spatiotemporal heterogeneity of potential OHCAs. To highlight the effectiveness of the proposed model, we applied the proposed method to Washington DC using user-generated anonymized mobile device location data. The results demonstrated that optimization-based planning provided an improved AED coverage level. We further evaluated the effectiveness of adding additional AEDs by analyzing the cost-coverage increment curve. In general, our framework provides a systematic approach for municipalities to integrate inclusive planning and budget-limited efficiency into their final decision-making. Given the high practicality and adaptability of the framework, the OSTO is highly amenable to different healthcare facilities’ deployment tasks with flexible demand and resource restraints.

Understanding the relationship between risk factors, geospatial patterns, and disease outcomes is essential in health geography research. These relationships can inform the implementation of healthcare and public health strategies to improve health outcomes. To accurately uncover such complex relationships, it is necessary to have a predictive model capable of integrating both health variables and spatial information to forecast health outcomes, along with a tool to interpret and reveal the patterns identified by this model. We developed a Spatial Counterfactual Explainable Deep Learning model (SpaCE), comprising a spatially explicit health outcome predictor and a prototype-guided counterfactual explanation. The SpaCE model unifies geospatial and health variables to improve predictions and generates hypothetical examples with minimal changes but opposite outcomes. Using these counterfactuals, SpaCE assesses the impact of each variable in different spatial contexts. We evaluated the model for predicting cardiac arrest survival outcomes. With a 0.682 AUCROC score, the SpaCE exceeds baseline models by 10.2%. Further analysis also reveals that the geospatial context significantly affects how various risk factors affect the survival outcomes of patients. Overall, the SpaCE model significantly improves predictive accuracy and explainability. It provides targeted interventions at both individual and geographic levels, and the cardiac arrest case study shows its high adaptability to various disease scenarios.