{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T17:50:29Z","timestamp":1791827429460,"version":"4.3.4"},"reference-count":84,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T00:00:00Z","timestamp":1791763200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    Accurate forecasting of fine particulate matter (PM\n                    <jats:sub>2.5<\/jats:sub>\n                    ) is critical for mitigating severe health risks in South Asia, yet it remains challenging in data-scarce regions like Bangladesh. While spatiotemporal graph neural networks (GNNs) offer a promising solution by capturing atmospheric transport, their effectiveness and the physical validity of their assumed spatial graphs have rarely been rigorously tested in such constrained settings. This study addresses this gap by evaluating a graph diffusion gated recurrent unit (GD-GRU) model using four years of daily, multi-source environmental data across a 12-city network. For forecasts up to three days ahead, the GD-GRU achieves an overall mean absolute error of 8.853 \u00b5g\/m\n                    <jats:sup>3<\/jats:sup>\n                    and an R\n                    <jats:sup>2<\/jats:sup>\n                    of 0.666, significantly outperforming non-graph baselines under FDR-corrected Diebold\u2013Mariano testing (p\u2009&lt;\u20090.001). Furthermore, the model significantly outperforms DCRNN when pooled across all forecast horizons (p\u2009&lt;\u20090.001), while remaining closely matched at individual lead times. Crucially, empirical validation of the model\u2019s underlying spatial assumptions demonstrates that occlusion-based spatial influence closely tracks the fixed geographic adjacency (r\u2009=\u20090.928, p\u2009&lt;\u20090.001), and this simple distance-based graph outperforms unconstrained adaptive, wind-informed, and hybrid alternatives. Finally, a multi-method explainability framework shows strong rank convergence (Spearman \u03c1\u2009=\u20090.590 to 0.783), identifying the Green Leaf Index, wind speed, atmospheric pressure, and seasonal encoding as the dominant predictors of PM\n                    <jats:sub>2.5<\/jats:sub>\n                    . These findings confirm that carefully validated, interpretable static graph models can deliver robust air quality forecasts without requiring dense sensor networks, offering a practical template for environmental monitoring in data-limited regions.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0351928","type":"journal-article","created":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T17:24:36Z","timestamp":1791825876000},"page":"e0351928","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Explainable spatiotemporal Graph Neural Networks: Multi-City PM2.5 forecasting in Bangladesh with geographic graph topology validation"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-3981-2127","authenticated-orcid":true,"given":"Pritom Ray","family":"Nobin","sequence":"first","affiliation":[{"name":"Khulna University of Engineering & Technology (KUET)","place":["Khulna, Bangladesh"],"department":["Department of Urban and Regional Planning"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,12]]},"reference":[{"key":"pone.0351928.ref001","doi-asserted-by":"crossref","first-page":"1223","DOI":"10.1016\/S0140-6736(20)30752-2","article-title":"Global burden of 87 risk factors in 204 countries and territories, 1990\u20132019: a systematic analysis for the Global Burden of Disease Study 2019","volume":"396","author":"CJ Murray","year":"2020","journal-title":"The Lancet"},{"key":"pone.0351928.ref002","unstructured":"2023 World Air Quality Report. Goldach, Switzerland: IQAir. 2024. https:\/\/www.iqair.com\/newsroom\/waqr-2023-pr"},{"key":"pone.0351928.ref003","author":"World Bank, Institute for Health Metrics and Evaluation. The Cost of Air Pollution: Strengthening the Economic Case for Action","year":"2016"},{"key":"pone.0351928.ref004","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1155\/2016\/8318453","article-title":"Investigating incursion of transboundary pollution into the atmosphere of Dhaka, Bangladesh","author":"MM Rana","year":"2016","journal-title":"Adv Meteorol"},{"issue":"2","key":"pone.0351928.ref005","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1007\/s11869-012-0187-2","article-title":"Particulate pollution from brick kiln clusters in the Greater Dhaka region, Bangladesh","volume":"6","author":"SK Guttikunda","year":"2012","journal-title":"Air Qual Atmos Health"},{"issue":"1","key":"pone.0351928.ref006","doi-asserted-by":"crossref","first-page":"205","DOI":"10.5194\/gmd-3-205-2010","article-title":"Incremental testing of the Community Multiscale Air Quality (CMAQ) modeling system version 4.7","volume":"3","author":"KM Foley","year":"2010","journal-title":"Geosci Model Dev"},{"key":"pone.0351928.ref007","doi-asserted-by":"crossref","DOI":"10.52202\/068431-0037","article-title":"Why do tree-based models still outperform deep learning on tabular data?","author":"L Grinsztajn","year":"2022"},{"key":"pone.0351928.ref008","doi-asserted-by":"crossref","first-page":"114798","DOI":"10.1016\/j.envpol.2020.114798","article-title":"Assessing the PM2.5 impact of biomass combustion in megacity Dhaka, Bangladesh","volume":"264","author":"MM Rahman","year":"2020","journal-title":"Environ Pollut"},{"key":"pone.0351928.ref009","doi-asserted-by":"crossref","first-page":"1401","DOI":"10.1016\/j.enceco.2025.07.001","article-title":"A systematic study on PM2.5 and PM10 concentration prediction in air pollution using machine learning and deep learning model","volume":"7","author":"P Patel","year":"2025","journal-title":"Environ Chem Ecotoxicol"},{"key":"pone.0351928.ref010","first-page":"15","article-title":"A spatiotemporal prediction framework for air pollution based on deep RNN","volume":"4","author":"J Fan","year":"2017","journal-title":"ISPRS Ann Photogramm Remote Sens Spat Inf Sci"},{"key":"pone.0351928.ref011","doi-asserted-by":"crossref","first-page":"13535","DOI":"10.1007\/s13762-023-04900-1","article-title":"Modeling air quality PM2.5 forecasting using deep sparse attention-based transformer networks","volume":"20","author":"Z Zhang","year":"2023","journal-title":"Int J Environ Sci Technol"},{"key":"pone.0351928.ref012","doi-asserted-by":"crossref","first-page":"106501","DOI":"10.1016\/j.envsoft.2025.106501","article-title":"A graph neural network and Transformer-based model for PM2.5 prediction through spatiotemporal correlation","volume":"191","author":"Y Ye","year":"2025","journal-title":"Environ Modell Software"},{"key":"pone.0351928.ref013","doi-asserted-by":"crossref","DOI":"10.1145\/3397536.3422208","article-title":"PM2.5-GNN: A domain knowledge enhanced graph neural network For PM2.5 forecasting","author":"S Wang","year":"2020"},{"key":"pone.0351928.ref014","author":"Y Li","year":"2017"},{"issue":"7","key":"pone.0351928.ref015","doi-asserted-by":"crossref","first-page":"3988","DOI":"10.3390\/ijerph19073988","article-title":"Regional prediction of ozone and fine particulate matter using diffusion convolutional recurrent neural network","volume":"19","author":"D Wang","year":"2022","journal-title":"Int J Environ Res Public Health"},{"key":"pone.0351928.ref016","first-page":"1","author":"PR Nobin","year":"2026"},{"key":"pone.0351928.ref017","author":"PR Nobin","year":"2025"},{"issue":"2","key":"pone.0351928.ref018","doi-asserted-by":"crossref","first-page":"100301","DOI":"10.1016\/j.geits.2025.100301","article-title":"A Fourier graph neural network for SOH estimation of lithium-ion batteries simultaneously considering spatio-temporal features","volume":"5","author":"W Liu","year":"2026","journal-title":"Green Energy and Intelligent Transportation"},{"issue":"5","key":"pone.0351928.ref019","doi-asserted-by":"crossref","first-page":"100267","DOI":"10.1016\/j.geits.2025.100267","article-title":"Lithium-ion battery remaining useful life prediction based on data-driven and particle filter fusion model","volume":"4","author":"C Wu","year":"2025","journal-title":"Green Energy Intelligent Transportation"},{"issue":"1","key":"pone.0351928.ref020","doi-asserted-by":"crossref","first-page":"17841","DOI":"10.1038\/s41598-024-68874-x","article-title":"Air quality forecasting using a spatiotemporal hybrid deep learning model based on VMD-GAT-BiLSTM","volume":"14","author":"X Wang","year":"2024","journal-title":"Sci Rep"},{"key":"pone.0351928.ref021","doi-asserted-by":"crossref","first-page":"121012","DOI":"10.1016\/j.atmosenv.2024.121012","article-title":"Enhanced PM2.5 prediction with a dynamic wind-driven graph fusion model incorporating inter-station pollutant transport","volume":"343","author":"H Gu","year":"2025","journal-title":"Atmos Environ"},{"key":"pone.0351928.ref022","doi-asserted-by":"crossref","first-page":"107971","DOI":"10.1016\/j.envint.2023.107971","article-title":"72-hour real-time forecasting of ambient PM2.5 by hybrid graph deep neural network with aggregated neighborhood spatiotemporal information","volume":"176","author":"M Teng","year":"2023","journal-title":"Environ Int"},{"key":"pone.0351928.ref023","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1016\/j.isprsjprs.2025.04.019","article-title":"SCAGAT: A scene-aware ensemble graph attention network for global PM2.5 pollution mapping via land\u2013atmosphere interactions","volume":"225","author":"K Bai","year":"2025","journal-title":"ISPRS Journal of Photogrammetry and Remote Sensing"},{"key":"pone.0351928.ref024","doi-asserted-by":"crossref","DOI":"10.1007\/s10462-026-11496-8","article-title":"Harnessing deep learning for air pollution forecasting: trends, techniques, and future prospects","volume":"59","author":"S Lawrence","year":"2026","journal-title":"Artif Intell Rev"},{"issue":"1","key":"pone.0351928.ref025","article-title":"Improving air quality assessment using physics-inspired deep graph learning","volume":"6","author":"L Li","year":"2023","journal-title":"npj Clim Atmos Sci"},{"key":"pone.0351928.ref026","author":"S Lundberg","year":"2017"},{"issue":"1","key":"pone.0351928.ref027","doi-asserted-by":"crossref","first-page":"230151","DOI":"10.4209\/aaqr.230151","article-title":"Interpretable Machine Learning Approaches for Forecasting and Predicting Air Pollution: A Systematic Review","volume":"24","author":"A Houdou","year":"2024","journal-title":"Aerosol Air Qual Res"},{"key":"pone.0351928.ref028","unstructured":"Molnar C. Interpretable Machine Learning: A Guide for Making Black Box Models Explainable. 2025."},{"key":"pone.0351928.ref029","author":"S Hooker","year":"2018","journal-title":"A Benchmark for Interpretability Methods in Deep Neural Networks"},{"issue":"10","key":"pone.0351928.ref030","doi-asserted-by":"crossref","first-page":"1247","DOI":"10.1007\/s11869-020-00878-8","article-title":"Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh","volume":"13","author":"SA Shahriar","year":"2020","journal-title":"Air Qual Atmos Health"},{"issue":"1","key":"pone.0351928.ref031","doi-asserted-by":"crossref","first-page":"220082","DOI":"10.4209\/aaqr.220082","article-title":"Spatio-temporal Variation of Meteorological Influence on PM2.5 and PM10 over Major Urban Cities of Bangladesh","volume":"23","author":"N Islam","year":"2023","journal-title":"Aerosol Air Qual Res"},{"key":"pone.0351928.ref032","unstructured":"Ministry of Environment, Forest and Climate Change G of the PR of B. Continuous Air Monitoring Station (CAMS) Network: Daily PM2.5 Monitoring Data for Major Urban Cities of Bangladesh. 2025. https:\/\/doe.gov.bd"},{"key":"pone.0351928.ref033","unstructured":"Zippenfenig P. Open-Meteo.com Weather API. Zenodo; 2024. https:\/\/doi.org\/10.5281\/zenodo.14582479\">10.5281\/zenodo.14582479"},{"key":"pone.0351928.ref034","doi-asserted-by":"crossref","first-page":"1999","DOI":"10.1002\/qj.3803","article-title":"The ERA5 global reanalysis","volume":"146","author":"H Hersbach","year":"2020","journal-title":"Q J R Meteorol Soc"},{"key":"pone.0351928.ref035","author":"E Vermote","year":"2021"},{"key":"pone.0351928.ref036","doi-asserted-by":"crossref","first-page":"6465","DOI":"10.1016\/j.atmosenv.2008.04.044","article-title":"An evaluation of Terra-MODIS sampling for monthly and annual particulate matter air quality assessment over the Southeastern United States","volume":"42","author":"P Gupta","year":"2008","journal-title":"Atmospheric Environment"},{"issue":"7569","key":"pone.0351928.ref037","doi-asserted-by":"crossref","first-page":"367","DOI":"10.1038\/nature15371","article-title":"The contribution of outdoor air pollution sources to premature mortality on a global scale","volume":"525","author":"J Lelieveld","year":"2015","journal-title":"Nature"},{"key":"pone.0351928.ref038","doi-asserted-by":"crossref","first-page":"8889","DOI":"10.5194\/acp-15-8889-2015","article-title":"Tropospheric ozone and its precursors from the urban to the global scale from air quality to short-lived climate forcer","volume":"15","author":"PS Monks","year":"2015","journal-title":"Atmos Chem Phys"},{"key":"pone.0351928.ref039","article-title":"The shuttle radar topography mission","volume":"45","author":"TG Farr","year":"2007","journal-title":"Rev Geophys"},{"issue":"2","key":"pone.0351928.ref040","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0107042","article-title":"Disaggregating census data for population mapping using random forests with remotely-sensed and ancillary data","volume":"10","author":"FR Stevens","year":"2015","journal-title":"PLoS One"},{"key":"pone.0351928.ref041","doi-asserted-by":"crossref","first-page":"6","DOI":"10.4324\/9781003169246-1","article-title":"VIIRS night-time lights","author":"CD Elvidge","year":"2021","journal-title":"Remote Sensing of Night-time Light"},{"key":"pone.0351928.ref042","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1016\/j.atmosenv.2014.11.049","article-title":"Imputation of missing data in time series for air pollutants","volume":"102","author":"WL Junger","year":"2015","journal-title":"Atmos Environ"},{"key":"pone.0351928.ref043","first-page":"158","article-title":"Virtues of the Haversine","volume":"68","author":"RW Sinnott","year":"1984","journal-title":"Sky Telesc"},{"key":"pone.0351928.ref044","author":"JP Snyder","year":"1987"},{"key":"pone.0351928.ref045","doi-asserted-by":"crossref","first-page":"1373","DOI":"10.1162\/089976603321780317","article-title":"Laplacian eigenmaps for dimensionality reduction and data representation","volume":"15","author":"M Belkin","year":"2003","journal-title":"Neural Comput"},{"issue":"3","key":"pone.0351928.ref046","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1109\/MSP.2012.2235192","article-title":"The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains","volume":"30","author":"DI Shuman","year":"2013","journal-title":"IEEE Signal Process Mag"},{"key":"pone.0351928.ref047","doi-asserted-by":"crossref","DOI":"10.2307\/j.ctt7t8hg","author":"DJ Jacob","year":"1999","journal-title":"Introduction to Atmospheric Chemistry"},{"key":"pone.0351928.ref048","volume-title":"Atmospheric chemistry and physics: From air pollution to climate change","author":"JH Seinfeld","year":"2016","edition":"3"},{"key":"pone.0351928.ref049","doi-asserted-by":"crossref","DOI":"10.1007\/978-94-009-3027-8","volume-title":"An introduction to boundary layer meteorology","author":"RB Stull","year":"1988"},{"key":"pone.0351928.ref050","doi-asserted-by":"crossref","DOI":"10.1090\/cbms\/092","author":"F Chung","year":"1996","journal-title":"Spectral graph theory"},{"key":"pone.0351928.ref051","author":"M Defferrard","year":"2016"},{"key":"pone.0351928.ref052","unstructured":"Rong Y, Huang W, Xu T, Huang J. DropEdge: Towards deep graph convolutional networks on node classification. In: Proceedings of the 8th International Conference on Learning Representations. 2020. https:\/\/openreview.net\/forum?id=Hkx1qkrKPr"},{"key":"pone.0351928.ref053","doi-asserted-by":"crossref","first-page":"117857","DOI":"10.1016\/j.oceaneng.2024.117857","article-title":"Deep learning reconstruction of high-Reynolds-number turbulent flow field around a cylinder based on limited sensors","volume":"304","author":"R Li","year":"2024","journal-title":"Ocean Engineering"},{"key":"pone.0351928.ref054","article-title":"Generalizable super-resolution turbulence reconstruction from minimal training data","volume":"1024","author":"H Wu","year":"2025","journal-title":"J Fluid Mech"},{"issue":"9","key":"pone.0351928.ref055","doi-asserted-by":"crossref","first-page":"3848","DOI":"10.1109\/TITS.2019.2935152","article-title":"T-GCN: A Temporal graph convolutional network for traffic prediction","volume":"21","author":"L Zhao","year":"2020","journal-title":"IEEE Trans Intell Transport Syst"},{"key":"pone.0351928.ref056","first-page":"1724","author":"K Cho","year":"2014"},{"key":"pone.0351928.ref057","author":"JL Ba","year":"2016"},{"key":"pone.0351928.ref058","author":"D Hendrycks","year":"2016"},{"key":"pone.0351928.ref059","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"N Srivastava","year":"2014","journal-title":"J Mach Learn Res"},{"key":"pone.0351928.ref060","author":"I Sutskever","year":"2014"},{"issue":"2","key":"pone.0351928.ref061","doi-asserted-by":"crossref","first-page":"270","DOI":"10.1162\/neco.1989.1.2.270","article-title":"A learning algorithm for continually running fully recurrent neural networks","volume":"1","author":"RJ Williams","year":"1989","journal-title":"Neural Computation"},{"issue":"1","key":"pone.0351928.ref062","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1214\/aoms\/1177703732","article-title":"Robust Estimation of a Location Parameter","volume":"35","author":"PJ Huber","year":"1964","journal-title":"Ann Math Statist"},{"key":"pone.0351928.ref063","author":"I Loshchilov","year":"2017"},{"issue":"1","key":"pone.0351928.ref064","doi-asserted-by":"crossref","first-page":"108","DOI":"10.1162\/neco.1995.7.1.108","article-title":"Training with noise is equivalent to tikhonov regularization","volume":"7","author":"CM Bishop","year":"1995","journal-title":"Neural Computation"},{"issue":"3","key":"pone.0351928.ref065","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1109\/JPROC.2021.3060483","article-title":"Explaining deep neural networks and beyond: A review of methods and applications","volume":"109","author":"W Samek","year":"2021","journal-title":"Proc IEEE"},{"key":"pone.0351928.ref066","author":"M Ancona","year":"2017"},{"key":"pone.0351928.ref067","unstructured":"Shrikumar A, Greenside P, Kundaje A. Learning Important Features Through Propagating Activation Differences. Proceedings of the 34th International Conference on Machine Learning. PMLR; 2017. pp. 3145\u201353."},{"key":"pone.0351928.ref068","author":"K Simonyan","year":"2013"},{"key":"pone.0351928.ref069","first-page":"177","article-title":"All models are wrong, but many are useful: Learning a variable\u2019s importance by studying an entire class of prediction models simultaneously","volume":"20","author":"A Fisher","year":"2019","journal-title":"J Mach Learn Res"},{"key":"pone.0351928.ref070","author":"A Duval","year":"2021"},{"key":"pone.0351928.ref071","doi-asserted-by":"crossref","first-page":"818","DOI":"10.1007\/978-3-319-10590-1_53","article-title":"Visualizing and understanding convolutional networks","author":"MD Zeiler","year":"2014","journal-title":"Lecture Notes in Computer Science"},{"key":"pone.0351928.ref072","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"L Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"pone.0351928.ref073","first-page":"785","author":"T Chen","year":"2016"},{"key":"pone.0351928.ref074","doi-asserted-by":"crossref","first-page":"253","DOI":"10.1080\/07350015.1995.10524599","article-title":"Comparing predictive accuracy","volume":"13","author":"FX Diebold","year":"1995","journal-title":"J Bus Econ Stat"},{"key":"pone.0351928.ref075","doi-asserted-by":"crossref","first-page":"281","DOI":"10.1016\/S0169-2070(96)00719-4","article-title":"Testing the equality of prediction mean squared errors","volume":"13","author":"D Harvey","year":"1997","journal-title":"Int J Forecast"},{"key":"pone.0351928.ref076","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"A Paszke","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"pone.0351928.ref077","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pone.0351928.ref078","doi-asserted-by":"crossref","first-page":"266","DOI":"10.3390\/atmos17030266","article-title":"Nonlinear impacts of air pollutants and meteorological factors on PM2.5: An interpretable GT-iFormer model with SHAP analysis","volume":"17","author":"D Li","year":"2026","journal-title":"Atmosphere"},{"key":"pone.0351928.ref079","article-title":"Characterization of PM2.5, gaseous pollutants, and meteorological interactions in the context of time-series health effects models","author":"K Ito","year":"2007","journal-title":"J Expo Sci Environ Epidemiol"},{"key":"pone.0351928.ref080","doi-asserted-by":"crossref","first-page":"137036","DOI":"10.1016\/j.jclepro.2023.137036","article-title":"A city-based PM2.5 forecasting framework using spatially attentive cluster-based graph neural network model","volume":"405","author":"S Mandal","year":"2023","journal-title":"J Clean Prod"},{"key":"pone.0351928.ref081","author":"JR Holton","year":"2013","journal-title":"An Introduction to Dynamic Meteorol"},{"key":"pone.0351928.ref082","doi-asserted-by":"crossref","first-page":"7705","DOI":"10.1016\/j.atmosenv.2010.10.022","article-title":"Key issues in controlling air pollutants in Dhaka, Bangladesh","volume":"45","author":"BA Begum","year":"2011","journal-title":"Atmospheric Environment"},{"key":"pone.0351928.ref083","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1016\/S0034-4257(02)00096-2","article-title":"Overview of the radiometric and biophysical performance of the MODIS vegetation indices","volume":"83","author":"A Huete","year":"2002","journal-title":"Remote Sensing of Environment"},{"key":"pone.0351928.ref084","author":"A Perotti","year":"2022"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0351928","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T17:24:46Z","timestamp":1791825886000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0351928"}},"subtitle":[],"editor":[{"given":"Lei","family":"Zhang","sequence":"first","affiliation":[{"name":"Beijing Institute of Technology","place":["CHINA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,12]]},"references-count":84,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,12]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0351928","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,12]]}}}