{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T17:42:23Z","timestamp":1792777343158,"version":"4.5.0"},"reference-count":49,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T00:00:00Z","timestamp":1792713600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000030","name":"Centers for Disease Control and Prevention","doi-asserted-by":"publisher","award":["CDC-RFA-FT-23-0069"],"award-info":[{"award-number":["CDC-RFA-FT-23-0069"]}],"id":[{"id":"10.13039\/100000030","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>The (large) season-to-season variability and limited dynamical history make the forecasting of infectious diseases a challenging problem. Here, we examine the extent to which advances in data-driven dynamical modeling can provide accurate predictions by benchmarking the performance of one such method, Hankel dynamic mode decomposition (DMD), on the 2024\u22122025 influenza, respiratory syncytial virus (RSV), and COVID-19 seasons in the United States and Canada. Using Hankel-DMD, we generated weekly forecasts that were submitted to the Center for Disease Control and Prevention\u2019s (CDC) FluSight Forecast Hub and the University of Guelph\u2019s AI4CastingHub. Across both Hubs, we find that Hankel-DMD can provide high quality forecasts at the beginning and end of the season, but forecasts in the middle of the season suffer from significant overestimation of the season peak. This leads to worse than baseline performance on FluSight Forecast Hub, when submissions are evaluated across the entire season. Hankel-DMD is found to be the best among a small number of models that submitted enough forecasts against which to compare for forecasting influenza, RSV, and COVID-19 in Canada. As this was the first active year of AI4CastingHub, this suggests that Hankel-DMD may be especially useful when expertise is lacking for predicting infectious dynamics in new regions. To study how Hankel-DMD can be better equipped to provide accurate forecasts during the season peak, we explored whether DMD with memory, a recently developed approach for applying DMD to non-stationary dynamical systems, provides significant improvements. Retrospective analysis using DMD with memory shows promise in reducing overestimation during season peak, but the extent to which its optimal use can be achieved in a live forecasting setting remains to be seen. Collectively, our results demonstrate that Hankel-DMD has potential as an efficient and interpretable method for the forecasting of infectious diseases, while also highlighting a number of remaining methodological challenges which future work should aim to address.<\/jats:p>","DOI":"10.1371\/journal.pone.0360068","type":"journal-article","created":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T17:36:40Z","timestamp":1792777000000},"page":"e0360068","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Data-driven forecasting of Flu, RSV, and COVID-19 related outcomes in the United States and Canada via Hankel dynamic mode decomposition"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4147-2026","authenticated-orcid":true,"given":"William T.","family":"Redman","sequence":"first","affiliation":[{"name":"Intelligent Systems Center, Johns Hopkins Applied Physics Laboratory","place":["Laurel, Maryland, United States of America"]},{"name":"Johns Hopkins University","place":["Baltimore, Maryland, United States of America"],"department":["Electrical and Computer Engineering Department"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4668-9803","authenticated-orcid":true,"given":"Luke C.","family":"Mullany","sequence":"additional","affiliation":[{"name":"Intelligent Systems Center, Johns Hopkins Applied Physics Laboratory","place":["Laurel, Maryland, United States of America"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,23]]},"reference":[{"key":"pone.0360068.ref001","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1016\/j.epidem.2017.08.002","article-title":"The RAPIDD ebola forecasting challenge: Synthesis and lessons learnt","volume":"22","author":"C Viboud","year":"2018","journal-title":"Epidemics"},{"issue":"11","key":"pone.0360068.ref002","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1007486","article-title":"Accuracy of real-time multi-model ensemble forecasts for seasonal influenza in the U.S","volume":"15","author":"NG Reich","year":"2019","journal-title":"PLoS Comput Biol"},{"issue":"8","key":"pone.0360068.ref003","doi-asserted-by":"crossref","first-page":"3146","DOI":"10.1073\/pnas.1812594116","article-title":"A collaborative multiyear, multimodel assessment of seasonal influenza forecasting in the United States","volume":"116","author":"NG Reich","year":"2019","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"1","key":"pone.0360068.ref004","doi-asserted-by":"crossref","first-page":"683","DOI":"10.1038\/s41598-018-36361-9","article-title":"Collaborative efforts to forecast seasonal influenza in the United States, 2015\u20132016","volume":"9","author":"CJ McGowan","year":"2019","journal-title":"Sci Rep"},{"issue":"48","key":"pone.0360068.ref005","doi-asserted-by":"crossref","first-page":"24268","DOI":"10.1073\/pnas.1909865116","article-title":"An open challenge to advance probabilistic forecasting for dengue epidemics","volume":"116","author":"MA Johansson","year":"2019","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"1","key":"pone.0360068.ref006","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1038\/s41597-022-01517-w","article-title":"The United States COVID-19 Forecast Hub dataset","volume":"9","author":"EY Cramer","year":"2022","journal-title":"Sci Data"},{"key":"pone.0360068.ref007","doi-asserted-by":"crossref","DOI":"10.7554\/eLife.81916","article-title":"Predictive performance of multi-model ensemble forecasts of COVID-19 across European nations","volume":"12","author":"K Sherratt","year":"2023","journal-title":"Elife"},{"issue":"1","key":"pone.0360068.ref008","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1186\/s13071-022-05630-y","article-title":"Evaluation of an open forecasting challenge to assess skill of West Nile virus neuroinvasive disease prediction","volume":"16","author":"KM Holcomb","year":"2023","journal-title":"Parasit Vectors"},{"issue":"5","key":"pone.0360068.ref009","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1011200","article-title":"Challenges of COVID-19 Case Forecasting in the US, 2020\u20132021","volume":"20","author":"VK Lopez","year":"2024","journal-title":"PLoS Computat Biol"},{"issue":"1","key":"pone.0360068.ref010","doi-asserted-by":"crossref","first-page":"6289","DOI":"10.1038\/s41467-024-50601-9","article-title":"Title evaluation of FluSight influenza forecasting in the 2021-22 and 2022-23 seasons with a new target laboratory-confirmed influenza hospitalizations","volume":"15","author":"SM Mathis","year":"2024","journal-title":"Nat Commun"},{"issue":"6","key":"pone.0360068.ref011","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1006134","article-title":"Nonmechanistic forecasts of seasonal influenza with iterative one-week-ahead distributions","volume":"14","author":"LC Brooks","year":"2018","journal-title":"PLoS Comput Biol"},{"issue":"1","key":"pone.0360068.ref012","doi-asserted-by":"crossref","first-page":"2991","DOI":"10.1038\/s41467-021-23234-5","article-title":"Multiscale influenza forecasting","volume":"12","author":"D Osthus","year":"2021","journal-title":"Nat Commun"},{"issue":"8","key":"pone.0360068.ref013","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1004382","article-title":"Flexible Modeling of Epidemics with an Empirical Bayes Framework","volume":"11","author":"LC Brooks","year":"2015","journal-title":"PLoS Comput Biol"},{"issue":"2","key":"pone.0360068.ref014","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1006599","article-title":"Even a good influenza forecasting model can benefit from internet-based nowcasts, but those benefits are limited","volume":"15","author":"D Osthus","year":"2019","journal-title":"PLoS Comput Biol"},{"issue":"1","key":"pone.0360068.ref015","article-title":"Integrative COVID-19 biological network inference with probabilistic core decomposition","volume":"23","author":"Y Guo","year":"2022","journal-title":"Brief Bioinform"},{"key":"pone.0360068.ref016","doi-asserted-by":"crossref","first-page":"03029","DOI":"10.7189\/jogh.13.03029","article-title":"Predicting the daily counts of COVID-19 infection using temporal convolutional networks","volume":"13","author":"M Li","year":"2023","journal-title":"J Glob Health"},{"issue":"6176","key":"pone.0360068.ref017","doi-asserted-by":"crossref","first-page":"1203","DOI":"10.1126\/science.1248506","article-title":"Big data. The parable of Google Flu: traps in big data analysis","volume":"343","author":"D Lazer","year":"2014","journal-title":"Science"},{"key":"pone.0360068.ref018","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1016\/j.epidem.2018.02.003","article-title":"Results from the second year of a collaborative effort to forecast influenza seasons in the United States","volume":"24","author":"M Biggerstaff","year":"2018","journal-title":"Epidemics"},{"key":"pone.0360068.ref019","doi-asserted-by":"crossref","first-page":"101035","DOI":"10.1016\/j.eclinm.2021.101035","article-title":"The impact of lockdown timing on COVID-19 transmission across US counties","volume":"38","author":"X Huang","year":"2021","journal-title":"EClinicalMedicine"},{"key":"pone.0360068.ref020","doi-asserted-by":"crossref","first-page":"100580","DOI":"10.1016\/j.epidem.2022.100580","article-title":"Gecko: A time-series model for COVID-19 hospital admission forecasting","volume":"39","author":"MJ Panaggio","year":"2022","journal-title":"Epidemics"},{"key":"pone.0360068.ref021","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1017\/S0022112009992059","article-title":"Spectral analysis of nonlinear flows","volume":"641","author":"CW Rowley","year":"2009","journal-title":"J Fluid Mech"},{"issue":"6","key":"pone.0360068.ref022","doi-asserted-by":"crossref","first-page":"1307","DOI":"10.1007\/s00332-015-9258-5","article-title":"A Data\u2013Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition","volume":"25","author":"MO Williams","year":"2015","journal-title":"J Nonlinear Sci"},{"issue":"4","key":"pone.0360068.ref023","doi-asserted-by":"crossref","first-page":"2096","DOI":"10.1137\/17M1125236","article-title":"Ergodic Theory, Dynamic Mode Decomposition, and Computation of Spectral Properties of the Koopman Operator","volume":"16","author":"H Arbabi","year":"2017","journal-title":"SIAM J Appl Dyn Syst"},{"issue":"1","key":"pone.0360068.ref024","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1038\/s41467-017-00030-8","article-title":"Chaos as an intermittently forced linear system","volume":"8","author":"SL Brunton","year":"2017","journal-title":"Nat Commun"},{"issue":"5","key":"pone.0360068.ref025","doi-asserted-by":"crossref","first-page":"315","DOI":"10.1073\/pnas.17.5.315","article-title":"Hamiltonian Systems and Transformation in Hilbert Space","volume":"17","author":"BO Koopman","year":"1931","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pone.0360068.ref026","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1007\/s11071-005-2824-x","article-title":"Spectral Properties of Dynamical Systems, Model Reduction and Decompositions","volume":"41","author":"I Mezi\u0107","year":"2005","journal-title":"Nonlinear Dyn"},{"issue":"4","key":"pone.0360068.ref027","doi-asserted-by":"crossref","first-page":"047510","DOI":"10.1063\/1.4772195","article-title":"Applied Koopmanism","volume":"22","author":"M Budisi\u0107","year":"2012","journal-title":"Chaos"},{"issue":"1","key":"pone.0360068.ref028","doi-asserted-by":"crossref","first-page":"2090","DOI":"10.1038\/s41467-020-15582-5","article-title":"Data-driven analysis and forecasting of highway traffic dynamics","volume":"11","author":"AM Avila","year":"2020","journal-title":"Nat Commun"},{"issue":"1","key":"pone.0360068.ref029","doi-asserted-by":"crossref","first-page":"16313","DOI":"10.1038\/s41598-020-73211-z","article-title":"Exponentially decaying modes and long-term prediction of sea ice concentration using Koopman mode decomposition","volume":"10","author":"J Hogg","year":"2020","journal-title":"Sci Rep"},{"key":"pone.0360068.ref030","unstructured":"Cao AB, Planer M, Hogg J, Fonoberova M, Bookman L, Macdonald T. Tipping-Points and Robustness of Sea Ice using Koopman Mode Decomposition. 2023."},{"key":"pone.0360068.ref031","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jneumeth.2015.10.010","article-title":"Extracting spatial-temporal coherent patterns in large-scale neural recordings using dynamic mode decomposition","volume":"258","author":"BW Brunton","year":"2016","journal-title":"J Neurosci Methods"},{"key":"pone.0360068.ref032","doi-asserted-by":"crossref","unstructured":"Bistrian D, Dimitriu G, Navon I. Modeling dynamic patterns from COVID-19 data using randomized dynamic mode decomposition in predictive mode and ARIMA. In: AIP Conference Proceedings. vol. 2302. AIP Publishing LLC; 2020. 080002 p.","DOI":"10.1063\/5.0033963"},{"issue":"1","key":"pone.0360068.ref033","doi-asserted-by":"crossref","first-page":"5788","DOI":"10.1038\/s41598-024-55798-9","article-title":"A Koopman operator-based prediction algorithm and its application to COVID-19 pandemic and influenza cases","volume":"14","author":"I Mezi\u0107","year":"2024","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0360068.ref034","doi-asserted-by":"crossref","first-page":"011102","DOI":"10.1063\/5.0245244","article-title":"Koopman learning with episodic memory","volume":"35","author":"WT Redman","year":"2025","journal-title":"Chaos"},{"issue":"2","key":"pone.0360068.ref035","doi-asserted-by":"crossref","first-page":"247","DOI":"10.3934\/jcd.2015005","article-title":"A kernel-based method for data-driven koopman spectral analysis","volume":"2","author":"MO Williams","year":"2015","journal-title":"JCD"},{"key":"pone.0360068.ref036","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1017\/S0022112010001217","article-title":"Dynamic mode decomposition of numerical and experimental data","volume":"656","author":"PJ Schmid","year":"2010","journal-title":"J Fluid Mech"},{"issue":"10","key":"pone.0360068.ref037","doi-asserted-by":"crossref","first-page":"103111","DOI":"10.1063\/1.4993854","article-title":"Extended dynamic mode decomposition with dictionary learning: A data-driven adaptive spectral decomposition of the Koopman operator","volume":"27","author":"Q Li","year":"2017","journal-title":"Chaos"},{"key":"pone.0360068.ref038","first-page":"366","article-title":"Detecting strange attractors in turbulence.","volume-title":"Dynamical systems and turbulence, Warwick 1980: proceedings of a symposium held at the University of Warwick 1979\/80","author":"F Takens","year":"2006"},{"issue":"18","key":"pone.0360068.ref039","doi-asserted-by":"crossref","first-page":"690","DOI":"10.1016\/j.ifacol.2016.10.246","article-title":"Koopman mode decomposition for periodic\/quasi-periodic time dependence","volume":"49","author":"I Mezic","year":"2016","journal-title":"IFAC-PapersOnLine"},{"issue":"4","key":"pone.0360068.ref040","doi-asserted-by":"crossref","first-page":"2478","DOI":"10.1137\/17M1133610","article-title":"Koopman operator family spectrum for nonautonomous systems","volume":"17","author":"S Macesic","year":"2018","journal-title":"SIAM J Appl Dyn Syst"},{"issue":"1","key":"pone.0360068.ref041","doi-asserted-by":"crossref","first-page":"909","DOI":"10.1137\/16M1062296","article-title":"Generalizing Koopman Theory to Allow for Inputs and Control","volume":"17","author":"JL Proctory","year":"2018","journal-title":"SIAM J Appl Dyn Syst"},{"issue":"3","key":"pone.0360068.ref042","doi-asserted-by":"crossref","first-page":"1586","DOI":"10.1137\/18M1192329","article-title":"Online Dynamic Mode Decomposition for Time-Varying Systems","volume":"18","author":"H Zhang","year":"2019","journal-title":"SIAM J Appl Dyn Syst"},{"key":"pone.0360068.ref043","doi-asserted-by":"crossref","first-page":"117159","DOI":"10.1109\/ACCESS.2023.3326412","article-title":"Non-Stationary Dynamic Mode Decomposition","volume":"11","author":"J Ferr\u00e9","year":"2023","journal-title":"IEEE Access"},{"key":"pone.0360068.ref044","unstructured":"2024-2025 Flu Forecasting Evaluation Report. Available from: https:\/\/www.cdc.gov\/flu-forecasting\/evaluation\/2024-2025-report.html"},{"issue":"2","key":"pone.0360068.ref045","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1008618","article-title":"Evaluating epidemic forecasts in an interval format","volume":"17","author":"J Bracher","year":"2021","journal-title":"PLoS Comput Biol"},{"key":"pone.0360068.ref046","unstructured":"2023-2024 Flu Forecasting Evaluation Report. Available from: https:\/\/www.cdc.gov\/flu-forecasting\/evaluation\/2023-2024-report.html"},{"key":"pone.0360068.ref047","doi-asserted-by":"crossref","first-page":"12271","DOI":"10.52202\/075280-0538","article-title":"Koopa: Learning non-stationary time series dynamics with koopman predictors","volume":"36","author":"Y Liu","year":"2023","journal-title":"Adv Neural Inform Process Syst"},{"issue":"2","key":"pone.0360068.ref048","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1137\/15M1023543","article-title":"Multiresolution Dynamic Mode Decomposition","volume":"15","author":"JN Kutz","year":"2016","journal-title":"SIAM J Appl Dyn Syst"},{"issue":"1","key":"pone.0360068.ref049","doi-asserted-by":"crossref","first-page":"4268","DOI":"10.1038\/s41467-024-48033-6","article-title":"Revealing trends and persistent cycles of non-autonomous systems with autonomous operator-theoretic techniques","volume":"15","author":"G Froyland","year":"2024","journal-title":"Nat Commun"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0360068","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T17:36:53Z","timestamp":1792777013000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0360068"}},"subtitle":[],"editor":[{"given":"Hans A.","family":"Kestler","sequence":"first","affiliation":[{"name":"University of Ulm","place":["GERMANY"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,23]]},"references-count":49,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,23]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0360068","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,23]]}}}