{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T07:16:33Z","timestamp":1788419793064,"version":"build-2803163510"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Public Health"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Influenza-like illness (ILI) poses a significant burden on public health systems, particularly in densely populated megacities. Accurate and timely forecasting is crucial for early warning and resource allocation. While deep learning models have shown promise, capturing the complex, non-linear coupling between environmental drivers and human behavior remains a challenge.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      We used weekly ILI% surveillance data from Beijing (2010\u20132019), complemented by 20 meteorological variables and 322 Baidu Search Keyword Indices. Cross-correlation function (CCF) analysis with seasonal differencing, ADF stationarity testing, and two-stage Benjamini\u2013Hochberg FDR correction (\n                      <jats:italic>q<\/jats:italic>\n                      &amp;lt; 0.05) identified 11 predictive features, 6 meteorological variables and 5 search-derived features. An Adapted DeformTime model (DeformTime-Ad) was developed, incorporating a task-decoupled attention block, epidemic-level sample weighting, per-fold Optuna hyperparameter optimization, and Huber loss. Performance was evaluated against 10 comparators: persistence, seasonal-na\u00efve, calendar-based, SARIMA, ARIMAX, SARIMAX, XGBoost, Random Forest, LSTM, and unadapted DeformTime-Base, under a 3-fold rolling-origin protocol across four horizons (7\u201328 days). Pairwise Diebold\u2013Mariano tests, stratified evaluation by epidemic intensity, and six ablation variants were conducted.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Beijing ILI% exhibited a pronounced unimodal winter\u2013spring peak. On common target weeks, DeformTime-Ad achieved the lowest MAE at 7-\u221214 days (0.39\u20130.45) and the lowest MSE at 7\u201321 days, representing up to 62.0% MSE reduction over a reporting-lag-corrected persistence benchmark. In Diebold\u2013Mariano tests, DeformTime-Ad was significantly superior to 8, 6, 2, and 1 of 10 comparators at 7, 14, 21, and 28 days, respectively, with zero significant losses at any horizon. During epidemic peaks (5 common target weeks), DeformTime-Ad attained a peak MAE of 2.09, 31.9% lower than DeformTime-Base, though the persistence benchmark achieved the lowest peak MAE (0.78) on this small sample, reflecting high autocorrelation during sustained peak plateaus. Ablation analysis identified per-fold hyperparameter optimization as the most impactful component (+51.9% MSE when removed), followed among architectural components by the task-decoupled attention block (+7.9%); epidemic-level sample weighting improved peak-period accuracy at a small overall-MSE trade-off. Meteorological and search-index features provided complementary, horizon-dependent predictive signals.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>A multimodal deep-learning architecture with epidemic-aware design substantially improves weekly ILI% forecasting in Beijing, particularly at 1\u20133 week horizons. Both meteorological and internet-search features contribute predictive signal in a horizon-dependent manner, with meteorological variables gaining prominence at longer lead times. The framework, combining exploratory CCF-based feature screening, leakage-free rolling-origin evaluation on common target weeks, and component-level ablation, offers a methodology warranting further prospective, multi-city validation before operational deployment for infectious disease surveillance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpubh.2026.1856514","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T06:42:19Z","timestamp":1788417739000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Fusing meteorological factors and baidu search index with an adapted deformtime model to predict influenza-like illness in Beijing"],"prefix":"10.3389","volume":"14","author":[{"given":"Lingling","family":"Shen","sequence":"first","affiliation":[{"name":"Beijing Meteorological Data Center","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yajun","family":"Xiong","sequence":"additional","affiliation":[{"name":"Beijing Meteorological Data Center","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhaobin","family":"Sun","sequence":"additional","affiliation":[{"name":"Chinese Academy of Meteorological Sciences","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"key":"B1","year":"2018"},{"key":"B2","doi-asserted-by":"publisher","first-page":"122","DOI":"10.2471\/BLT.17.194514","article-title":"Revision of clinical case definitions: influenza-like illness and severe acute respiratory infection","volume":"96","year":"2018","journal-title":"Bull World Health Organ"},{"key":"B3","doi-asserted-by":"crossref","first-page":"439","DOI":"10.1289\/ehp.1002383","volume":"119","year":"2011","journal-title":"Environ Health Perspect"},{"key":"B4","doi-asserted-by":"publisher","first-page":"116171","DOI":"10.1016\/j.envres.2023.116171","article-title":"Spatiotemporal effects of meteorological conditions on global influenza peaks","volume":"231","author":"Liang","year":"2023"},{"key":"B5","doi-asserted-by":"publisher","first-page":"e14406","DOI":"10.1371\/journal.pone.0009450","article-title":"Modeling and predicting seasonal influenza transmission in warm regions using climatological parameters","volume":"5","author":"Soebiyanto","year":"2010","journal-title":"PLoS One"},{"key":"B6","doi-asserted-by":"publisher","first-page":"135678","DOI":"10.1016\/j.jhazmat.2024.136114","article-title":"Temperature variability and influenza incidence in China: Effect modification by ambient fine particulate matter","volume":"480","author":"Li","year":"2024","journal-title":"J Hazard Mater"},{"key":"B7","doi-asserted-by":"publisher","first-page":"e1001552","DOI":"10.1371\/journal.pmed.1001552","article-title":"Characterization of regional influenza seasonality patterns in China and implications for vaccination strategies: spatio-temporal modeling of surveillance data","volume":"10","author":"Yu","year":"2013","journal-title":"PLoS Med"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1695","DOI":"10.1093\/infdis\/jix181","article-title":"Effect of weather variability on seasonal influenza among different age groups in Queensland, Australia: a Bayesian spatiotemporal analysis","volume":"215","author":"Huang","year":"2017","journal-title":"J Infect Dis"},{"key":"B9","doi-asserted-by":"publisher","first-page":"e0134701","DOI":"10.1371\/journal.pone.0134701","article-title":"Associations between meteorological parameters and influenza activity in Berlin (Germany), Ljubljana (Slovenia), Castile and Le\u00f3n (Spain) and Israeli districts","volume":"10","author":"Soebiyanto","year":"2015","journal-title":"PLoS One"},{"key":"B10","author":"Zhong","year":"2020","journal-title":"A study of influenza epidemic characteristics, influencing factors and model prediction in China [Doctoral dissertation]. Guangzhou: Guangzhou University of Chinese Medicine"},{"key":"B11","doi-asserted-by":"publisher","first-page":"151724","DOI":"10.1016\/j.scitotenv.2021.151724","article-title":"Influenza seasonality and its environmental driving factors in mainland China and Hong Kong","volume":"818","author":"Ali","year":"2022","journal-title":"Sci Total Environ"},{"key":"B12","doi-asserted-by":"publisher","first-page":"e02426","DOI":"10.1128\/AEM.02426-18","article-title":"Effects of absolute humidity, relative humidity, temperature, and wind speed on influenza activity in Toronto, Ontario, Canada","volume":"85","author":"Peci","year":"2019","journal-title":"Appl Environ Microbiol"},{"key":"B13","doi-asserted-by":"publisher","first-page":"157024","DOI":"10.1016\/j.scitotenv.2022.157024","volume":"843","author":"Zhang","year":"2023","journal-title":"Sci Total Environ"},{"key":"B14","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1093\/infdis\/jis526","article-title":"Epidemiological dynamics and phylogeography of influenza virus in southern China","volume":"207","author":"Cheng","year":"2013","journal-title":"J Infect Dis"},{"key":"B15","doi-asserted-by":"publisher","first-page":"17534","DOI":"10.1038\/s41598-017-17806-z","article-title":"Seasonal pattern of influenza activity in a subtropical city, China, 2010\u20132015","volume":"7","author":"Liu","year":"2017","journal-title":"Sci Rep"},{"key":"B16","author":"Shan","year":"2025","journal-title":"Analysis of influenza epidemic characteristics and research on dynamic trend prediction methods in China from 2009 to 2018 [Doctoral dissertation]. Beijing: Academy of Military Sciences"},{"key":"B17","doi-asserted-by":"publisher","first-page":"304","DOI":"10.1111\/zph.13114","article-title":"Forecasting seasonal influenza activity in Canada-Comparing seasonal Auto-Regressive integrated moving average and artificial neural network approaches for public health preparedness","volume":"71","author":"Orang","year":"2024","journal-title":"Zoonoses Public Health"},{"key":"B18","doi-asserted-by":"publisher","first-page":"629","DOI":"10.46234\/ccdcw2024.084","article-title":"Exploring the lagged correlation between Baidu Index and influenza-like illness \u2013 China, 2014\u20132019","volume":"6","author":"Han","year":"2024","journal-title":"China CDC Wkly"},{"key":"B19","doi-asserted-by":"publisher","first-page":"e1004513","DOI":"10.1371\/journal.pcbi.1004513","article-title":"Combining search, social media, and traditional data sources to improve influenza surveillance","volume":"11","author":"Santillana","year":"2015","journal-title":"PLoS Comput Biol"},{"key":"B20","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1186\/s12889-025-25937-6","article-title":"Dynamic ensemble deep learning with multi-source data for robust influenza forecasting in Yangzhou","volume":"25","author":"Wang","year":"2025","journal-title":"BMC Public Health"},{"key":"B21","doi-asserted-by":"publisher","first-page":"206","DOI":"10.3772\/j.issn.1000-0135.2018.02.009","article-title":"Mechanism and empirical research on forecasting influenza epidemic fused with Baidu Index","volume":"37","author":"Wang","year":"2018","journal-title":"J China Soc Sci Tech Inf"},{"key":"B22","doi-asserted-by":"publisher","first-page":"e44238","DOI":"10.2196\/44238","article-title":"Deep-learning model for influenza prediction from multisource heterogeneous data in a megacity: model development and evaluation","volume":"25","author":"Yang","year":"2023","journal-title":"J Med Internet Res"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1858","DOI":"10.3390\/ijerph19031858","article-title":"The prediction of influenza-like illness and respiratory disease using LSTM and ARIMA","volume":"19","author":"Tsan","year":"2022","journal-title":"Int J Environ Res Public Health"},{"key":"B24","doi-asserted-by":"publisher","first-page":"3249","DOI":"10.1038\/s41467-021-23440-1","article-title":"Impact of COVID-19 outbreaks and interventions on influenza in China and the United States","volume":"12","author":"Feng","year":"2021","journal-title":"Nat Commun"},{"key":"B25","article-title":"DeformTime: capturing variable dependencies with deformable attention for time series forecasting","author":"Shu","year":"2025"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1170","DOI":"10.1128\/JCM.42.3.1170-1175.2004","article-title":"Computer-assisted surveillance for detecting clonal outbreaks of nosocomial infection","volume":"42","author":"Hacek","year":"2004","journal-title":"J Clin Microbiol"},{"key":"B27","author":"Box","year":"2015"},{"key":"B28","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"B29","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"B30","first-page":"18","article-title":"Classification and regression by randomForest","volume":"2","author":"Liaw","year":"2002","journal-title":"R News"},{"key":"B31","doi-asserted-by":"publisher","first-page":"785","DOI":"10.1145\/2939672.2939785","article-title":"XGBoost: A scalable tree boosting system","author":"Chen","year":"2016"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1776470","DOI":"10.3389\/fpubh.2026.1776470","article-title":"Incorporating meteorological factors into a SARIMA model for predicting pediatric influenza epidemics","volume":"14","author":"Mu","year":"2026","journal-title":"Front Public Health"},{"key":"B33","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1186\/s41182-025-00723-7","article-title":"Assessing dengue forecasting methods: a comparative study of statistical models and machine learning techniques in Rio de Janeiro, Brazil","volume":"53","author":"Chen","year":"2025","journal-title":"Trop Med Health"},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-306-47630-3","article-title":"Principles of Forecasting: A Handbook for Researchers and Practitioners","author":"Armstrong","year":"2001"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1247","DOI":"10.5194\/gmd-7-1247-2014","article-title":"Root mean square error (RMSE) or mean absolute error (MAE)?","volume":"7","author":"Chai","year":"2014","journal-title":"Arguments against avoiding RMSE in the literature. Geosci Model Dev"},{"key":"B36","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1016\/j.idm.2024.04.010","article-title":"Predicting influenza-like illness trends based on sentinel surveillance data in China from 2011 to 2019: a modeling and comparative study","volume":"9","author":"Zhang","year":"2024","journal-title":"Infect Dis Model"},{"key":"B37","doi-asserted-by":"publisher","first-page":"e45085","DOI":"10.2196\/45085","article-title":"Influenza epidemic trend surveillance and prediction based on search engine data: deep learning model study","volume":"25","author":"Yang","year":"2023","journal-title":"J Med Internet Res"},{"key":"B38","article-title":"GB\/T 20484-2017 Grade of cold air","year":"2017"},{"key":"B39","doi-asserted-by":"publisher","first-page":"527","DOI":"10.1186\/s12889-026-26753-2","article-title":"Modeling the dynamics of seasonal influenza in response to meteorological conditions and antigenic variation: a simulation study","volume":"26","author":"Liang","year":"2026","journal-title":"BMC Public Health"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1856514\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T06:42:23Z","timestamp":1788417743000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1856514\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":39,"alternative-id":["10.3389\/fpubh.2026.1856514"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1856514","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]},"article-number":"1856514"}}