Pith. sign in

REVIEW 4 major objections 5 minor 89 references

SatHealth: A Multimodal Public Health Dataset with Satellite-based Environmental Factors

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read SatHealth builds a public Ohio dataset pairing satellite images and environmental variables with claims-derived disease prevalence, and argues that adding regional environment embeddings improves AI models' accuracy and spatiotemporal…

desk verdict A genuinely new and useful dataset for environmental health AI; the 'significantly improve' claim is not statistically supported as reported. read the letter →

arxiv 2506.13842 v1 pith:Y7XXYCBV submitted 2025-06-16 cs.LG

Yuanlong Wang , Pengqi Wang , Changchang Yin , Ping Zhang This is my paper · ORCID
classification cs.LG
keywords satelliteimageryenvironmentalhealthinformaticssocialdeterminantsofdiseaseprevalencemedicalclaimsmultimodaldataspatiotemporalgeneralizationregionalenvironmentembeddings
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

SatHealth is a public dataset that brings the living environment into AI health research: for every Ohio county, ZIP-code tabulation area, census tract, and metropolitan area from 2016 to 2022, it joins satellite images, climate and air-quality time series, land cover, a social-deprivation index, and yearly prevalence estimates for 1,377 ICD-coded diseases computed from medical claims. The load-bearing claim is that this environmental information is not decoration: adding it to models improves prediction of regional social-deprivation scores and disease prevalence, and also improves patient-level next-visit and one-year disease-risk prediction across several sequence models. The paper also reports that environment-augmented models generalize better under spatial interpolation, spatial extrapolation, and temporal forecasting, and it releases the embeddings so other models can use them without reprocessing raw imagery. If the claim is right, environment-aware AI can be built on top of existing electronic-health-record backbones rather than treated as a separate data collection burden.

What carries the argument

The regional environment embedding is the object that carries the argument. Dynamic monthly variables (climate, air quality, greenery) are averaged by meteorological season and concatenated into an annual vector; static land-cover fractions and pixel-statistics features derived from satellite images (RGB channels plus nine vegetation and color indices) are added to form a multimodal yearly profile for each region. These embeddings are then fed to random-forest regressors for regional health outcomes and concatenated to patient representations from EHR sequence models, with an optional boosting step that adds neighborhood and historical residual predictors. The same embedding object is what turns raw geospatial data into a plug-in health-AI input.

What would settle it

Using the same Ohio CBSAs and years, compare SatHealth's ICD prevalence estimates with an external population-representative source such as CDC PLACES or Ohio state public health data; then train the paper's regional and patient-level models to predict the external target instead of the MarketScan label. If the environment embeddings stop improving accuracy against the external outcomes, or the urban-rural prevalence gaps disappear, the central claim that living environmental information improves health AI is an artifact of the claims population rather than a general result.

Watch

Extended reading notes

Core claim

The paper's central discovery is a reusable multimodal public resource plus the evidence that it works: SatHealth fuses high-resolution Google Maps satellite imagery (432,918 images, each about 500 meters square), monthly environmental variables (climate, air quality, greenery), static land cover, and 2019 Social Deprivation Index values with disease prevalences estimated from 2,141,777 Ohio patients in the MarketScan commercial claims database. From these features the authors construct regional environment embeddings at four geographic levels, then use them in two tasks. On regional modeling, combining all modalities raises $R^2$ for neoplasm prevalence by 0.086 over the best single-modality features and matches or beats single modalities for diabetes and hypertension targets. On personalized prediction, adding the environmental embedding to LSTM, RETAIN, Dipole, and Transformer backbones improves macro AUROC and recall metrics for most settings; with Dipole, next-visit mAUC rises from 0.600 to 0.722. The authors claim this makes SatHealth the first US dataset to combine regional environmental characteristics with a healthcare database, and they interpret the gains as evidence that living environmental information can significantly improve model performance and temporal-spatial generalizability.

Load-bearing premise

The health labels are not population-representative: disease prevalence is computed from the MarketScan commercial claims database, which covers insured employees and their dependents, and the paper does not validate these estimates against CDC surveys, Medicare, or state registries; if that insured population differs from Ohio as a whole, the environment-health correlations and prediction gains are systematically biased.

Editorial extensions

If this is right

  • Researchers can add environmental context to any health model with a patient or region location by using SatHealth's precomputed embeddings, avoiding raw satellite and weather processing.
  • The reported results imply that EHR-based risk prediction is leaving signal on the table when it ignores residence: the largest gains appear on recall, meaning environment helps surface diseases a model would otherwise miss.
  • Combining dynamic and static modalities seems more useful than any single modality, so future SatHealth-style resources should keep the multimodal design rather than simplify to one data source.
  • The spatiotemporal-enhancement results suggest that models trained on this dataset should include neighborhood and history features when deployed across regions or years.
  • The published pipeline and web application allow construction of the same dataset for other US states, which the authors say will be updated toward national coverage.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the MarketScan prevalence estimates are not representative of Ohio's general population, the observed urban-rural odds ratios could partly reflect who is insured and which providers bill claims, so the environmental associations should be checked against population-representative surveys before being treated as public-health facts.
  • The paper's own Limitations section notes that only Ohio is covered, that the embeddings are simple feature-engineered statistics rather than learned representations, and that patient residence is coarse; these bound the generalizability claim but do not undermine the resource itself.
  • Because the embeddings are computed from public geospatial inputs alone, they could plausibly be reused for tasks the paper does not study, such as hospital-resource planning or environmental-justice screening, but those extensions would need their own outcome validation.
  • The correlation analyses suggest testable causal hypotheses (green space and soil conditions linked to cardiovascular and metabolic disease), but the dataset's cross-sectional design cannot by itself distinguish environment effects from population sorting.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. This manuscript introduces SatHealth, a multimodal public health dataset for Ohio (2016–2022) that combines satellite imagery, climate/air-quality/greenery time series, land cover fractions, Social Deprivation Index (SDI) scores, and disease prevalences estimated from the MarketScan CCAE commercial claims database. The authors propose a deterministic embedding pipeline that fuses these modalities into regional environmental embeddings and then evaluate the dataset in two use cases: regional regression of SDI and disease prevalence, and personalized disease-risk prediction with LSTM, RETAIN, Dipole, and Transformer backbones. A further set of experiments examines spatial interpolation, spatial extrapolation, and temporal forecasting. The central claim is that living-environment information 'significantly improve[s] AI models' performance and temporal-spatial generalizability' (Abstract, Section 4). The paper also describes a web application and a public code repository for data access and reproduction.

Significance. If the central claim were established, SatHealth would be a valuable public resource: it is, to my knowledge, the first US dataset combining regional environmental characteristics with a claims-based all-disease prevalence panel, and the authors provide a web application, downloadable data, and a reproducible pipeline. The two use cases are sensible and the deterministic embedding approach is transparent and easy to reuse. However, the headline claim of significant improvement is currently under-supported: the prevalence targets are derived from an insured commercial population with unknown representativeness, the spatial granularity yields only about fifteen regional units, and the reported R2 differences are not accompanied by any significance testing, confidence intervals, or repeated-seed variability. These issues are load-bearing because the same experimental evidence is used in the abstract, Section 4, and the conclusion.

major comments (4)
  1. [§3.4, §6] The disease prevalence targets are estimated only from MarketScan CCAE, which covers employees and dependents in employer-sponsored plans, yet no external validation against population-representative sources (e.g., CDC surveys, Medicare, or state registries) is provided. Since prevalence is the health outcome in both use cases, systematic bias in the enrolled population would propagate into the environment-health correlations and the regression and prediction results. Section 6 lists limitations about Ohio-only coverage, simple embeddings, and coarse residence, but does not acknowledge or mitigate this representativeness concern, which is a material omission.
  2. [§4.2, Table 4] The claim that environmental information 'significantly' improves predictions is not statistically supported. The regional regressions use about 15 MSA/CBSA-level units and 7 years (on the order of 105 region-year observations) with hundreds of input features, and the reported R2 gains of 0.02–0.09 are not accompanied by confidence intervals, hypothesis tests, or repeated cross-validation with different seeds. Without such evidence, the improvements over single modalities may reflect noise or overfitting rather than genuine environmental signal. The authors should add significance testing or clearly reframe the claim as descriptive rather than inferential.
  3. [§4.3, Table 5] The benefit of environmental information is not uniform, which contradicts the blanket statement of significant improvement. For example, for RETAIN, next-visit Recall@50 decreases from 0.456 to 0.445 when environment is added; for LSTM, 1-year diagnosis Recall@5 decreases from 0.503 to 0.489 and Recall@10 from 0.537 to 0.508. The paper should report per-model, per-metric variability and should temper the general claim accordingly, distinguishing settings where the environmental embedding helps from those where it does not.
  4. [§4.4, Table 6] The spatiotemporal generalization results do not consistently support the claimed generalizability benefit. In the spatial extrapolation scenario, SDoH R2 is near zero or negative for every feature set (e.g., DEnv: -0.186, All: 0.035), and adding temporal information sometimes degrades performance (e.g., DEnv+T SDoH: -0.301 vs. DEnv: -0.186). These patterns should be discussed as evidence that the proposed spatiotemporal enhancement is not universally beneficial, and the claims in the text should be aligned with the full set of results rather than only the 'best or second best' cases.
minor comments (5)
  1. [§4.1.1] In the sentence referencing Table 3, 'Tabel' should be 'Table'.
  2. [§3.2, Table 11] 'Greenary' appears as a typo for 'Greenery' in several places, including Section 3.2 and Table 11.
  3. [Table 2] Table 2 lists 'SDI 8 Variables' while Section 3.1 and Appendix B.2.3 describe seven demographic components; the table should clarify whether the eighth variable is the overall SDI score.
  4. [§5] The web-application section would benefit from screenshots with readable text, and from stating whether the downloadable embeddings include the exact versions used in the reported experiments.
  5. [Appendix B.2.2] The statement that Google Maps images lack timestamps and are treated as static is important, but the assumption that the landscape is stable over 2016–2022 should be stated more explicitly as an assumption in the main text, not only in the appendix.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SatHealth's environmental features and health targets are independent, and the embeddings are deterministic.

full rationale

SatHealth's derivation chain separates sources: environmental variables and satellite images are collected from Earth observation products and Google Maps (Sections 3.2-3.3), while targets—SDI from ACS-based SDI scores (Section 3.1) and disease prevalence from MarketScan claims (Section 3.4)—come from independent databases. The regional embeddings are computed by spatial averaging, seasonal aggregation, and pixel-level statistics without fitting any target-derived parameters, so the 'All' feature set used in Tables 4-6 is not constructed from the labels it predicts. Personalized risk prediction concatenates these deterministic environmental embeddings to patient representations, and the patient-level labels are individual ICD codes, not the regional prevalence aggregates; no fitted parameter is renamed as a prediction. The only self-citation ([13], used as a pointer to prior MarketScan-based EHR modeling) is not load-bearing; the MarketScan source itself is independently cited ([45]). The paper's limitations and the reviewer's statistical concerns—no confidence intervals or significance tests for the 'significantly improve' claim, MarketScan's insured-population representativeness, and possible spatial autocorrelation—are external-validity or evidentiary issues, not cases where an equation reduces to its own input. The correlation analyses (odds ratios, Spearman correlations) are descriptive validations of the dataset rather than derivations that presuppose the health-environment relationship. Therefore, the central claims are not circular.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim rests on several domain assumptions: MarketScan claims are treated as a valid proxy for population disease prevalence; Google Maps images are assumed stable over the study period; the 2019 SDI is applied to all years; and regional averaging of environmental variables is assumed to capture exposure. These are not derived in the paper and are the main validity risks.

free parameters (2)
  • History decay factor
    Exponential decay factor for aggregating historical environmental features in the spatiotemporal enhanced regression (Section C.1.2); value not reported, chosen by hand.
  • Neighborhood aggregation coefficients
    Constant coefficients for aggregating neighborhood environmental features in the spatiotemporal enhanced regression (Section C.1.2); values not reported.
assumptions (4)
  • domain assumption MarketScan CCAE claims provide an unbiased proxy for regional disease prevalence.
    Used in Section 3.4 to estimate all-disease prevalence from 2.14M insured patients; no external validation against CDC or survey prevalence. The database excludes uninsured and most public-insurance populations.
  • domain assumption Google Maps satellite imagery is temporally stable over 2016-2022.
    Images have no timestamps; paper states this 'doesn't matter if we assume the stability of the landscape in several years' (Section B.2.2).
  • domain assumption The 2019 Social Deprivation Index is constant across 2016-2022.
    SDI only available for 2019 (Section A.1) but used as yearly SDoH target for all years in Table 7.
  • domain assumption Regional averaging of environmental variables preserves relevant environmental exposure signal.
    Spatial reduction by averaging pixels within regions (Section 3.2); assumes ecological exposure rather than individual-level exposure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SatHealth: A Multimodal Public Health Dataset with Satellite-based Environmental Factors." pith.science (2026). https://pith.science/paper/Y7XXYCBV

@misc{pith2026250613842,
  author       = {Pith},
  title        = {Pith review of: SatHealth: A Multimodal Public Health Dataset with Satellite-based Environmental Factors},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y7XXYCBV}},
  note         = {Machine review of arXiv:2506.13842}
}
read the original abstract

Living environments play a vital role in the prevalence and progression of diseases, and understanding their impact on patient's health status becomes increasingly crucial for developing AI models. However, due to the lack of long-term and fine-grained spatial and temporal data in public and population health studies, most existing studies fail to incorporate environmental data, limiting the models' performance and real-world application. To address this shortage, we developed SatHealth, a novel dataset combining multimodal spatiotemporal data, including environmental data, satellite images, all-disease prevalences estimated from medical claims, and social determinants of health (SDoH) indicators. We conducted experiments under two use cases with SatHealth: regional public health modeling and personal disease risk prediction. Experimental results show that living environmental information can significantly improve AI models' performance and temporal-spatial generalizability on various tasks. Finally, we deploy a web-based application to provide an exploration tool for SatHealth and one-click access to both our data and regional environmental embedding to facilitate plug-and-play utilization. SatHealth is now published with data in Ohio, and we will keep updating SatHealth to cover the other parts of the US. With the web application and published code pipeline, our work provides valuable angles and resources to include environmental data in healthcare research and establishes a foundational framework for future research in environmental health informatics.

Figures

Figures reproduced from arXiv: 2506.13842 by the authors.

Figure 1
Figure 1. Overview of SatHealth and Experimental framework. (A) The components of SatHealth. (B) The regional embedding [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Feature correlations - second level ICD the prevalence of each ICD-10 code for MSAs in a given year by cal￾culating the percentage of patients associated with the code among all patients recorded in that year. Data Processing. The ICD-10 codes are organized hierarchi￾cally, allowing for a multi-level investigation of disease prevalence. For instance, the code "I10" represents essential hypertension, its parent node … view at source ↗
Figure 2
Figure 2. Note that this figure only shows subsets of ICD codes and [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figures from the paper (4 more)
Figure 3
Figure 3. Figure 3: Random Forest Regression Performance from Dif [PITH_FULL_IMAGE:figures/full_fig_p008_3.png]
Figure 4
Figure 4. Figure 4: UI design of the proposed dataset environmental features in a bar chart. The correlation score and corresponding p-value will also be displayed below the bar chart. Users can search for a specific variable to check its correlation to the current disease. 6 Limitations …
Figure 5
Figure 5. Figure 5: Feature correlations - top level ICD [PITH_FULL_IMAGE:figures/full_fig_p017_5.png]
Figure 6
Figure 6. Figure 6: Feature correlations - third level ICD [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

89 extracted references · 67 canonical work pages

  1. [1]

    2024.Sentinel-2

    European Space Agency. 2024.Sentinel-2. https://sentiwiki.copernicus.eu/web/ sentinel-2

    work page 2024

  2. [2]

    Suraiya P Ahmad, Omar Torres, Pawan K Bhartia, Gregory Leptoukh, and SJ Kempler. 2006. Aerosol index from TOMS and OMI measurements. InProc. of the 86th AMS annual meeting

    work page 2006

  3. [3]

    Araújo, Daniel C

    Leonardo D. Araújo, Daniel C. Zanotta, Nicolas Ray, and Maurício R. Veronez

    work page

  4. [4]

    Aji Kusumaning Asri, Chia-Pin Yu, Wen-Chi Pan, Yue Leon Guo, Huey-Jen Su, Shih-Chun Candice Lung, Chih-Da Wu, and John D Spengler. 2020. Global greenness in relation to reducing the burden of cardiovascular diseases: ischemic heart disease and stroke.Environmental Research Letters15, 12 (Dec. 2020), 124003. doi:10.1088/1748-9326/abbbaf

    work page doi:10.1088/1748-9326/abbbaf 2020

  5. [5]

    Thomas Astell-Burt and Xiaoqi Feng. 2019. Urban green space, tree canopy, and prevention of heart disease, hypertension, and diabetes: a longitudinal study.The Lancet Planetary Health3 (Sept. 2019), S16. doi:10.1016/S2542-5196(19)30159-7 Publisher: Elsevier

    work page doi:10.1016/s2542-5196(19)30159-7 2019

  6. [6]

    Jannah Baker, Nicole White, and Kerrie Mengersen. 2014. Missing in space: an evaluation of imputation methods for missing data in spatial analysis of risk factors for type II diabetes.International Journal of Health Geographics13, 1 (Nov. 2014), 47. doi:10.1186/1476-072X-13-47

    work page doi:10.1186/1476-072x-13-47 2014

  7. [7]

    Evelise Pereira Barboza, Marta Cirach, Sasha Khomenko, Tamara Iungman, Na- talie Mueller, Jose Barrera-Gómez, David Rojas-Rueda, Michelle Kondo, and Mark Nieuwenhuijsen. 2021. Green space and mortality in European cities: a health impact assessment study.The Lancet Planetary Health5, 10 (Oct. 2021), e718–e730. doi:10.1016/S2542-5196(21)00229-1

    doi:10.1016/s2542-5196(21)00229-1 2021

  8. [8]

    Bhabananda Biswas, Fangjie Qi, Jayanta Kumar Biswas, Ayanka Wijayawardena, Muhammad Atikul Islam Khan, and Ravi Naidu. 2018. The Fate of Chemical Pollutants with Soil Properties and Processes in the Climate Change Paradigm—A Review.Soil Systems2, 3 (Sept. 2018), 51. doi:10.3390/soilsystems2030051 Number: 3 Publisher: Multidisciplinary Digital Publishing Institute

    work page doi:10.3390/soilsystems2030051 2018

Show all 89 references
  1. [9]

    2024.American Community Survey (ACS)

    US Census Bureau. 2024.American Community Survey (ACS). https://www. census.gov/programs-surveys/acs Section: Government

    2024

  2. [10]

    Xiaofeng Cao, Yulin Liu, Rui Yu, Dejun Han, and Baofeng Su. 2021. A Comparison of UAV RGB and Multispectral Imaging in Phenotyping for Stay Green of Wheat Population.Remote Sensing13, 24 (Dec. 2021), 5173. doi:10.3390/rs13245173

    2021 doi

  3. [11]

    CDC. 2024. About Rural Health. Retrieved 2025-01-16 from https://www.cdc. gov/rural-health/php/about/index.html

    2024

  4. [12]

    2024.ICD-10-CM

    CDC. 2024.ICD-10-CM. https://www.cdc.gov/nchs/icd/icd-10-cm/index.html

    2024

  5. [13]

    Jiayuan Chen, Changchang Yin, Yuanlong Wang, and Ping Zhang. 2024. Predictive Modeling with Temporal Graphical Representation on Electronic Health Records. doi:10.48550/arXiv.2405.03943 arXiv:2405.03943 [cs]

    2024 doi arXiv

  6. [14]

    Zhuo Chen, Jean-Eudes Dazard, Yassin Khalifa, Issam Motairek, Sadeer Al-Kindi, and Sanjay Rajagopalan. 2024. Artificial intelligence–based assessment of built environment from Google Street View and coronary artery disease prevalence. European Heart Journal45, 17 (May 2024), 1...

    2024 doi

  7. [15]

    Zhuo Chen, Jean-Eudes Dazard, Yassin Khalifa, Issam Motairek, Catherine Kreatsoulas, Sanjay Rajagopalan, and Sadeer Al-Kindi. 2024. Deep Learn- ing–Based Assessment of Built Environment From Satellite Images and Car- diometabolic Disease Prevalence.JAMA Cardiology9, 6 (June 20...

    2024

  8. [16]

    Stewart, and Jimeng Sun

    Edward Choi, Mohammad Taha Bahadori, Le Song, Walter F. Stewart, and Jimeng Sun. 2017. GRAM: Graph-based Attention Model for Healthcare Representation Learning. InProceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ’17). Asso...

    2017

  9. [17]

    Edward Choi, Mohammad Taha Bahadori, Jimeng Sun, Joshua Kulas, Andy Schuetz, and Walter Stewart. 2016. RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mech- anism. InAdvances in Neural Information Processing Systems, Vol. 29. Curran Associ...

    2016

  10. [18]

    Burton, Kelly S

    Anne Cleary, Anne Roiko, Nicola W. Burton, Kelly S. Fielding, Zoe Murray, and Gavin Turrell. 2019. Changes in perceptions of urban green space are related to changes in psychological well-being: Cross-sectional and longitudinal study of mid-aged urban residents.Health & Place5...

    2019

  11. [19]

    Coughlin, Catherine Clary, J

    Steven S. Coughlin, Catherine Clary, J. Aaron Johnson, Adam Berman, Vahe Heboyan, Teal Benevides, Justin Moore, and Varghese George. 2019. Continuing Challenges in Rural Health in the United States.Journal of environment and health sciences5, 2 (2019), 90–92. https://www.ncbi....

    2019

  12. [20]

    McEachan, Marta Cirach, Christopher J

    Payam Dadvand, John Wright, David Martinez, Xavier Basagaña, Rosemary R.C. McEachan, Marta Cirach, Christopher J. Gidlow, Kees De Hoogh, Regina Gražule- vičien˙e, and Mark J. Nieuwenhuijsen. 2014. Inequality, green spaces, and pregnant women: Roles of ethnicity and individual ...

    2014 doi

  13. [21]

    DiPietro Mager, Terrell W

    Natalie A. DiPietro Mager, Terrell W. Zollinger, Jack E. Turman, Jianjun Zhang, and Brian E. Dixon. 2021. Routine Healthcare Utilization Among Reproductive- Age Women Residing in a Rural Maternity Care Desert.Journal of Community Health46, 1 (Feb. 2021), 108–116. doi:10.1007/s...

    2021 doi

  14. [22]

    Haikuan Feng, Huilin Tao, Zhenhai Li, Guijun Yang, and Chunjiang Zhao. 2022. Comparison of UAV RGB Imagery and Hyperspectral Remote-Sensing Data for Monitoring Winter Wheat Growth.Remote Sensing14, 15 (Aug. 2022), 3811. doi:10.3390/rs14153811

    2022 doi

  15. [23]

    2024.Opioid and SUD Data Enclave (O-SUDDEn): Bringing real- time data to the opioid crisis

    Soledad Fernández. 2024.Opioid and SUD Data Enclave (O-SUDDEn): Bringing real- time data to the opioid crisis. https://reporter.nih.gov/project-details/10590246

    2024

  16. [24]

    Sebastien Garrigues, Julien Chimot, Melanie Ades, Antje Inness, Johannes Flem- ming, Zak Kipling, Istvan Laszlo, Angela Benedetti, Roberto Ribas, Soheila Ja- fariserajehlou, et al. 2022. Monitoring multiple satellite aerosol optical depth (AOD) products within the Copernicus A...

    2022

  17. [25]

    Nieuwenhuijsen

    Mireia Gascon, Margarita Triguero-Mas, David Martínez, Payam Dadvand, David Rojas-Rueda, Antoni Plasència, and Mark J. Nieuwenhuijsen. 2016. Residential green spaces and mortality: A systematic review.Environment International86 (Jan. 2016), 60–67. doi:10.1016/j.envint.2015.10.013

    2016 doi

  18. [26]

    Rory Gibb, Felipe J. Colón-González, Phan Trong Lan, Phan Thi Huong, Vu Sinh Nam, Vu Trong Duoc, Do Thai Hung, Nguyen Thanh Dong, Vien Chinh Chien, Ly Thi Thuy Trang, Do Kien Quoc, Tran Minh Hoa, Nguyen Huu Tai, Tran Thi Hang, Gina Tsarouchi, Eleanor Ainscoe, Quillon Harpham, ...

    2023

  19. [27]

    Goodchild and Wenwen Li

    Michael F. Goodchild and Wenwen Li. 2021. Replication across space and time must be weak in the social and environmental sciences.Proceedings of the National Academy of Sciences118, 35 (Aug. 2021), e2015759118. doi:10.1073/pnas. 2015759118

    2021 doi

  20. [28]

    2024.Google Maps Static API

    Google. 2024.Google Maps Static API. https://developers.google.com/maps/ documentation/maps-static/overview

    2024

  21. [29]

    Noel Gorelick, Matt Hancher, Mike Dixon, Simon Ilyushchenko, David Thau, and Rebecca Moore. 2017. Google Earth Engine: Planetary-scale geospatial analysis for everyone.Remote Sensing of Environment202 (Dec. 2017), 18–27. doi:10.1016/j.rse.2017.06.031

    2017 doi

  22. [30]

    Christelle Gée and Emmanuel Denimal. 2020. RGB Image-Derived Indicators for Spatial Assessment of the Impact of Broadleaf Weeds on Wheat Biomass.Remote Sensing12, 18 (Sept. 2020), 2982. doi:10.3390/rs12182982

    2020 doi

  23. [31]

    Kyle Heuton, Jyontika Kapoor, Shikhar Shrestha, Thomas J Stopka, and Michael C Hughes. 2024. Spatiotemporal Forecasting of Opioid-Related Fatal Overdoses: Towards Best Practices for Modeling and Evaluation.American Journal of Epi- demiology(Sept. 2024), kwae343. doi:10.1093/aj...

    2024 doi

  24. [32]

    Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long Short-Term Memory. Neural Comput.9, 8 (1997), 1735–1780. doi:10.1162/neco.1997.9.8.1735

    1997 doi

  25. [33]

    Pawinee Iamtrakul, Sararad Chayphong, Sajjakaj Jomnonkwao, and Vatanavongs Ratanavaraha. 2021. The Association of Falls Risk in Older Adults and Their Living Environment: A Case Study of Rural Area, Thailand.Sustainability13, 24 (Dec. 2021), 13756. doi:10.3390/su132413756

    2021 doi

  26. [34]

    Robert Graham Center Policy Studies in Family Medicine & Primary Care. 2018. Social Deprivation Index (SDI). Retrieved 2024-11-19 from https://www.graham- center.org/content/brand/rgc/maps-data-tools/social-deprivation-index.html

    2018

  27. [35]

    J Irizar, M Melf, P Bartsch, J Koehler, S Weiss, R Greinacher, M Erdmann, V Kirschner, A Perez Albinana, and D Martin. 2019. Sentinel-5/uvns. InInternational Conference on Space Optics—ICSO 2018, Vol. 11180. SPIE, 41–58

    2019

  28. [36]

    SHIGETO KAWASHIMA and MAKOTO NAKATANI. 1998. An Algorithm for Estimating Chlorophyll Content in Leaves Using a Video Camera.Annals of Botany81, 1 (Jan. 1998), 49–54. doi:10.1006/anbo.1997.0544

    1998

  29. [37]

    Mshana, Benon Asiimwe, John Kiiru, Andy G

    Katherine Keenan, Michail Papathomas, Stephen E. Mshana, Benon Asiimwe, John Kiiru, Andy G. Lynch, Mike Kesby, Stella Neema, Joseph R. Mwanga, Martha F. Mushi, Wei Jing, Dominique L. Green, Emmanuel Olamijuwon, Qing Zhang, Rachel Sippy, Kathryn J. Fredricks, Stephen H. Gillesp...

    2024

  30. [38]

    den Braver, and Jeroen Lakerveld

    Thao Minh Lam, Nicolette R. den Braver, and Jeroen Lakerveld. 2023. The Built Environment and Metabolic Syndrome. InMetabolic Syndrome: A Comprehensive SatHealth: A Multimodal Public Health Dataset with Satellite-based Environmental Factors KDD ’25, August 3–7, 2025, Toronto, ...

    2023

  31. [39]

    Xiaotong Li, Lihua Li, Wen Qin, Qing Cao, Xin Mu, Tiange Liu, Zhen Li, and Wei Zhang. 2023. Urban Living Environment and Myopia in Children.JAMA Network Open6, 12 (Dec. 2023), e2346999. doi:10.1001/jamanetworkopen.2023.46999

    2023

  32. [40]

    Loccoh, Maddox Karen E

    Eméfah C. Loccoh, Maddox Karen E. Joynt, Yun Wang, Dhruv S. Kazi, Robert W. Yeh, and Rishi K. Wadhera. 2022. Rural-Urban Disparities in Outcomes of Myocar- dial Infarction, Heart Failure, and Stroke in the United States.Journal of the Amer- ican College of Cardiology79, 3 (Jan...

    2022 doi

  33. [41]

    Jones, Zhihao Jin, Tamar Polonsky, Karen Kim, Christopher O

    Jiajun Luo, Rena R. Jones, Zhihao Jin, Tamar Polonsky, Karen Kim, Christopher O. Olopade, Jayant Pinto, Habibul Ahsan, and Briseis Aschebrook-Kilfoy. 2024. Dif- fering associations of PM2.5 exposure with systolic and diastolic blood pressures across exposure durations in a pre...

    2024 doi

  34. [42]

    Junyu Luo, Muchao Ye, Cao Xiao, and Fenglong Ma. 2020. HiTANet: Hierarchical Time-Aware Attention Networks for Risk Prediction on Electronic Health Records. InProceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. ACM, Virtual Event C...

    2020

  35. [43]

    Fenglong Ma, Radha Chitta, Jing Zhou, Quanzeng You, Tong Sun, and Jing Gao

  36. [44]

    Fenglong Ma, Quanzeng You, Houping Xiao, Radha Chitta, Jing Zhou, and Jing Gao. 2018. KAME: Knowledge-based Attention Model for Diagnosis Prediction in Healthcare. InProceedings of the 27th ACM International Conference on Information and Knowledge Management. ACM, Torino Italy...

    2018 doi

  37. [45]

    2024.Real world evidence

    Merative. 2024.Real world evidence. https://www.merative.com/real-world- evidence

    2024

  38. [46]

    Hanna Meyer and Edzer Pebesma. 2021. Predicting into unknown space? Esti- mating the area of applicability of spatial prediction models.Methods in Ecology and Evolution12, 9 (2021), 1620–1633. doi:10.1111/2041-210X.13650 _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1111...

    2021 doi

  39. [47]

    Lene Mikkelsen, Kim Moesgaard Iburg, Tim Adair, Thomas Fürst, Michael Heg- nauer, Elena von der Lippe, Lauren Moran, Shuhei Nomura, Haruka Sakamoto, Kenji Shibuya, Annelene Wengler, Stephanie Willbond, Patricia Wood, and Alan D. Lopez. 2020. Assessing the quality of cause of d...

    2020 doi

  40. [48]

    López, Lama Moukheiber, Mira Moukheiber, Sulaiman Moukheiber, Juan Sebastian Osorio- Valencia, Saptarshi Purkayastha, Atika Rahman Paddo, Chenwei Wu, and Po- Chih Kuo

    Dana Moukheiber, David Restrepo, Sebastián Andrés Cajas, María Patricia Ar- beláez Montoya, Leo Anthony Celi, Kuan-Ting Kuo, Diego M. López, Lama Moukheiber, Mira Moukheiber, Sulaiman Moukheiber, Juan Sebastian Osorio- Valencia, Saptarshi Purkayastha, Atika Rahman Paddo, Chenw...

    2024 doi

  41. [49]

    Muñoz Sabater

    J. Muñoz Sabater. 2024.ERA5-Land hourly data from 1950 to present. https: //cds.climate.copernicus.eu/datasets/reanalysis-era5-land?tab=overview

    2024

  42. [50]

    Nushrat Nazia, Jane Law, and Zahid Ahmad Butt. 2022. Identifying spatiotemporal patterns of COVID-19 transmissions and the drivers of the patterns in Toronto: a Bayesian hierarchical spatiotemporal modelling.Scientific Reports12, 1 (June 2022), 9369. doi:10.1038/s41598-022-13403-x

    2022 doi

  43. [51]

    Behnam Nikparvar, Md Mokhlesur Rahman, Faizeh Hatami, and Jean-Claude Thill. 2021. Spatio-temporal prediction of the COVID-19 pandemic in US counties: modeling with a deep LSTM neural network.Scientific Reports11, 1 (Nov. 2021), 21715. doi:10.1038/s41598-021-01119-3 Publisher:...

    2021 doi

  44. [52]

    Department of Agriculture (USDA)

    U.S. Department of Agriculture (USDA). 2025. Urban Soil Survey | Natural Re- sources Conservation Service. Retrieved 2025-01-21 from https://www.nrcs.usda. gov/conservation-basics/natural-resource-concerns/soil/urban-soil-survey

    2025

  45. [53]

    Department of Health, Office of Disease Prevention Human Services, and Health Promotion

    U.S. Department of Health, Office of Disease Prevention Human Services, and Health Promotion. 2024.Social Determinants of Health - Healthy People 2030 | odphp.health.gov. Retrieved 2024-11-19 from https://odphp.health.gov/ healthypeople/priority-areas/social-determinants-health

    2024

  46. [54]

    Patton, Wig Zamore, Elena N

    Allison P. Patton, Wig Zamore, Elena N. Naumova, Jonathan I. Levy, Doug Brugge, and John L. Durant. 2015. Transferability and Generalizability of Regression Models of Ultrafine Particles in Urban Neighborhoods in the Boston Area.Environ- mental Science & Technology49, 10 (May ...

    2015 doi

  47. [55]

    Muhammad Mainuddin Patwary, Mohammad Javad Zare Sakhvidi, Sadia Ashraf, Payam Dadvand, Matthew H. E. M. Browning, Md Ashraful Alam, Michelle L. Bell, Peter James, and Thomas Astell-Burt. 2024. Impact of green space and built environment on metabolic syndrome: A systematic revi...

    2024

  48. [56]

    Ricketts

    Thomas C. Ricketts. 2000. The Changing Nature of Rural Health Care.Annual Review of Public Health21, 1 (May 2000), 639–657. doi:10.1146/annurev.publhealth. 21.1.639

    2000 doi

  49. [57]

    Roberts, Volker Bahn, Simone Ciuti, Mark S

    David R. Roberts, Volker Bahn, Simone Ciuti, Mark S. Boyce, Jane Elith, Gurutzeta Guillera-Arroita, Severin Hauenstein, José J. Lahoz-Monfort, Boris Schröder, Wilfried Thuiller, David I. Warton, Brendan A. Wintle, Florian Hartig, and Carsten F. Dormann. 2017. Cross-validation ...

    2017 doi

  50. [58]

    Sanja Scepanovic, Ivica Obadic, Sagar Joglekar, Laura Giustarini, Cristiano Nat- tero, Daniele Quercia, and Xiaoxiang Zhu. 2024. MedSat: A Public Health Dataset for England Featuring Medical Prescriptions and Satellite Imagery.Advances in Neural Information Processing Systems36 (2024)

    2024

  51. [59]

    Jason Semprini, Khyathi Gadag, Gawain Williams, Aniyah Muldrow, and Whit- ney E. Zahnd. 2024. Rural–Urban Cancer Incidence and Trends in the United States, 2000 to 2019.Cancer Epidemiology, Biomarkers & Prevention33, 8 (Aug. 2024), 1012–1022. doi:10.1158/1055-9965.EPI-24-0072

    2024 doi

  52. [60]

    2024.Dynamic Land Cover

    Copernicus Land Monitoring Service. 2024.Dynamic Land Cover. https://land. copernicus.eu/en/products/global-dynamic-land-cover

    2024

  53. [61]

    Harvey Millar

    Santiago Signorelli, Esteban Casaretto, and A. Harvey Millar. 2023. Green Index: a widely accessible method to quantify greenness of photosynthetic organisms. doi:10.1101/2023.08.23.554481

    2023 doi

  54. [62]

    Adrian Spoerri, Marcel Zwahlen, Matthias Egger, and Matthias Bopp. 2010. The Swiss National Cohort: a unique database for national and international re- searchers.International Journal of Public Health55, 4 (Aug. 2010), 239–242. doi:10.1007/s00038-010-0160-5

    2010 doi

  55. [63]

    Kai Sun, Yingjie Hu, Gaurish Lakhanpal, and Ryan Zhenqi Zhou. 2023. Spatial cross-validation for GeoAI. InHandbook of Geospatial Artificial Intelligence. CRC Press, 201–214

    2023

  56. [64]

    Tzortzis, Maria Kaselimi, Ioannis Rallis, Anasta- sios Doulamis, and Nikolaos Doulamis

    Anastasios Temenos, Ioannis N. Tzortzis, Maria Kaselimi, Ioannis Rallis, Anasta- sios Doulamis, and Nikolaos Doulamis. 2022. Novel Insights in Spatial Epidemi- ology Utilizing Explainable AI (XAI) and Remote Sensing.Remote Sensing14, 13 (June 2022), 3074. doi:10.3390/rs14133074

    2022 doi

  57. [65]

    Stirling, and Arul Earnest

    Getayeneh Antehunegn Tesema, Zemenu Tadesse Tessema, Stephane Heritier, Rob G. Stirling, and Arul Earnest. 2023. A Systematic Review of Joint Spatial and Spatiotemporal Models in Health Research.International Journal of Environmental Research and Public Health20, 7 (March 2023...

    2023 doi

  58. [66]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is All you Need. (2017)

    2017

  59. [67]

    Wafula, Oumou Maiga-Ascofare, Nicole S

    Solomon T. Wafula, Oumou Maiga-Ascofare, Nicole S. Struck, Don P. Math- anga, Lauren M. Cohee, Jürgen May, Dewi I. Puradiredja, and Eva Lorenz. 2024. Socioeconomic disparities in Plasmodium falciparum infection risk in South- ern Malawi: mediation analyses.Scientific Reports14...

    2024 doi

  60. [68]

    Wei Wang, Liu Meng, Zheyu Hu, Xia Yuan, Weisi Zeng, Kunlun Li, Hanjia Luo, Min Tang, Xiao Zhou, Xiaoqiong Tian, Chenhui Luo, Yi He, and Shuo Yang. 2022. The association between outdoor air pollution and lung cancer risk in seven eastern metropolises of China: Trends in 2006-20...

    2022

  61. [69]

    Way, Jenny L

    Elora A. Way, Jenny L. Carwile, Erika C. Ziller, and Katherine A. Ahrens

  62. [70]

    Andrea Wendling, Valerie Taglione, Rachel Rezmer, Poe Lwin, Jessica Frost, Julia Terhune, and Jean Kerver. 2021. Access to maternity and prenatal care services in rural Michigan.Birth48, 4 (2021), 566–573. doi:10.1111/birt.12563 _eprint: https://onlinelibrary.wiley.com/doi/pdf...

    2021 doi

  63. [71]

    White, Ian Alcock, Benedict W

    Mathew P. White, Ian Alcock, Benedict W. Wheeler, and Michael H. Depledge

  64. [72]

    2025.Ill-defined diseases

    World Health Organization (WHO). 2025.Ill-defined diseases. Retrieved 2025- 01-16 from https://platform.who.int/mortality/themes/theme-details/MDB/ill- defined-diseases

    2025

  65. [73]

    Muhao Xu, Zhenfeng Zhu, Youru Li, Shuai Zheng, Yawei Zhao, Kunlun He, and Yao Zhao. 2024. FlexCare: Leveraging Cross-Task Synergy for Flexible Multimodal Healthcare Prediction. InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’24). As...

    2024

  66. [74]

    Zhichao Yang, Avijit Mitra, Weisong Liu, Dan Berlowitz, and Hong Yu. 2023. TransformEHR: transformer-based encoder-decoder generative model to en- hance prediction of disease outcomes using electronic health records.Nature Communications14, 1 (Nov. 2023), 7857. doi:10.1038/s41...

    2023 doi

  67. [75]

    Lobell, and Stefano Ermon

    Christopher Yeh, Chenlin Meng, Sherrie Wang, Anne Driscoll, Erik Rozi, Patrick Liu, Jihyeon Lee, Marshall Burke, David B. Lobell, and Stefano Ermon. 2021. SustainBench: Benchmarks for Monitoring the Sustainable Development Goals KDD ’25, August 3–7, 2025, Toronto, ON, Canada W...

    2021 arXiv

  68. [76]

    Ping Yin. 2019. Urban–rural inequalities in spatial accessibility to prenatal care: a GIS analysis of Georgia, USA, 2000–2010.GeoJournal84, 3 (June 2019), 671–683. doi:10.1007/s10708-018-9884-1

    2019 doi

  69. [77]

    Aiken, Richard Harris, and Jonathan L

    Xueqing Yin, John M. Aiken, Richard Harris, and Jonathan L. Bamber. 2024. A Bayesian spatio-temporal model of COVID-19 spread in England.Scientific Reports14, 1 (May 2024), 10335. doi:10.1038/s41598-024-60964-0 Publisher: Nature Publishing Group

    2024 doi

  70. [78]

    Zhihao Yu, Chaohe Zhang, Yasha Wang, Wen Tang, Jiangtao Wang, and Liantao Ma. 2024. Predict and Interpret Health Risk Using Ehr Through Typical Patients. InICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 1506–1510. doi:10.1...

    2024

  71. [79]

    Schenck, Zhenxing Xu, Yongkang Zhang, Jie Xu, Jiang Bian, Dmitry Morozyuk, Dhruv Khullar, Anna S

    Chengxi Zang, Yu Hou, Edward J. Schenck, Zhenxing Xu, Yongkang Zhang, Jie Xu, Jiang Bian, Dmitry Morozyuk, Dhruv Khullar, Anna S. Nordvig, Elizabeth A. Shenkman, Russell L. Rothman, Jason P. Block, Kristin Lyman, Yiye Zhang, Jay Varma, Mark G. Weiner, Thomas W. Carton, Fei Wan...

  72. [80]

    Hanze Zhang, Ningyuan Zhang, Zeyu Li, Jingyuan Yang, Yucheng Zhu, Zhao Liu, and Li Chen. 2024. A higher shadow ratio of the living environment on the remote sensing digital image is possibly protective for adolescent myopia. Scientific Reports14, 1 (Oct. 2024), 23824. doi:10.1...

    2024 doi

  73. [81]

    Hengbiao Zheng, Tao Cheng, Dong Li, Xiang Zhou, Xia Yao, Yongchao Tian, Weixing Cao, and Yan Zhu. 2018. Evaluation of RGB, Color-Infrared and Mul- tispectral Images Acquired from Unmanned Aerial Systems for the Estima- tion of Nitrogen Accumulation in Rice.Remote Sensing10, 6 ...

    2018 doi

  74. [82]

    Xue-yan Zheng, Si-li Tang, Tao Liu, Ye Wang, Xiao-jun Xu, Ni Xiao, Chuan Li, Yan-jun Xu, Zhao-xuan He, Shu-li Ma, Yu-liang Chen, Rui-lin Meng, and Li-feng Lin. 2022. Effects of long-term PM2.5 exposure on metabolic syndrome among adults and elderly in Guangdong, China.Environm...

    2022

  75. [84]

    fraction maps,

    doi:10.1186/s12940-022-00888-2 A Dataset Information A.1 Data Scope We started with data curation in Ohio while participating in the Ohio O-SUDDEn program [23]. We noticed the shortage of medi- cal datasets with environmental health risk factors. Therefore, we develop SatHealt...

    doi

  76. [85]

    doi:10.1038/s43856-024-00549-0

    Identification of risk factors of Long COVID and predictive modeling in the RECOVER EHR cohorts.Communications Medicine4, 1 (July 2024), 130. doi:10.1038/s43856-024-00549-0

    2024 doi

  77. [228]

    doi:10.1007/978-3-031-40116-9_59

    doi

  78. [2013]

    Would You Be Happier Living in a Greener Urban Area? A Fixed-Effects Analysis of Panel Data.Psychological Science24, 6 (June 2013), 920–928. doi:10. 1177/0956797612464659

    2013

  79. [2017]

    InProceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

    Dipole: Diagnosis Prediction in Healthcare via Attention-based Bidirec- tional Recurrent Neural Networks. InProceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, Halifax NS Canada, 1903–1911. doi:10.1145/3097983.3098088

    1903

  80. [2022]

    doi:10.1111/ppe.12862 _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1111/ppe.12862

    Out-of-hospital births and infant mortality in the United States: Ef- fect measure modification by rural maternal residence.Paediatric and Peri- natal Epidemiology36, 3 (2022), 399–411. doi:10.1111/ppe.12862 _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1111/ppe.12862

    2022 doi

  81. [2024]

    2024), 20933

    Earth observation data uncover green spaces’ role in mental health.Scientific Reports14, 1 (Sept. 2024), 20933. doi:10.1038/s41598-024-72008-8 Publisher: Nature Publishing Group

    2024 doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.