{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T19:01:01Z","timestamp":1783537261018,"version":"3.55.0"},"reference-count":51,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:00:00Z","timestamp":1783468800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Key Technologies R&D Program of China","award":["2023YFC2306700"],"award-info":[{"award-number":["2023YFC2306700"]}]}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>\n                    Accurate forecasting of infectious disease cases and deaths is crucial for public health decision-making. Traditional statistical and machine learning approaches may be challenged by complex temporal patterns and heterogeneous surveillance data. We evaluated large language models (LLMs) for infectious disease forecasting. We collected monthly reported cases and deaths data from China National Notifiable Diseases Surveillance System (NNDSS) from January 2009 to February 2025, and monthly reported cases from the United States NNDSS from 2016 to 2023, covering 79 notifiable infectious diseases and 111 forecasting tasks. We evaluated seven models: four statistical models (ARIMA, TGARCH, EGARCH, ETS), two machine learning models (XGBoost, LSTM), and an LLM-based regression model built on Qwen-2.5-3B and fine-tuned using Low-Rank Adaptation (LoRA). Performance was assessed using mean absolute error (MAE), root mean squared error (RMSE), and mean absolute percentage error (MAPE). We used Friedman tests and Nemenyi post-hoc comparisons versus the LLM, reporting mean-rank differences (\u0394R;\u2009&gt;\u20090 indicates better LLM ranking). In pooled analyses, the Friedman test indicated significant heterogeneity across models (\u03c7\u00b2\u2009=\u200926.24,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001). Nemenyi comparisons showed positive \u0394R for all baselines against the LLM, with significance only in the XGBoost\u2013LLM comparison (\u0394R\u2009=\u20090.72,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001). By category, significant positive \u0394R for all baselines against the LLM was observed in zoonotic infections (XGBoost: \u0394R\u2009=\u20091.42,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001) and intestinal infections (TGARCH: \u0394R\u2009=\u20091.28,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.012; EGARCH: \u0394R\u2009=\u20091.98,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001), whereas blood-borne infections showed LLM inferiority (ETS: \u0394R=\u22122.29,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.011; XGBoost \u0394R=\u22122.86,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001). Geographically, positive \u0394R was significant in China (XGBoost: \u0394R\u2009=\u20090.80,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.007), while the U.S. stratum was not significant (\u03c7\u00b2\u2009=\u20099.04,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.171). By outcome, between-model differences were detectable for case-forecasting (\u03c7\u00b2\u2009=\u200920.72,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.002; XGBoost: \u0394R\u2009=\u20090.73,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.005), with no detectable differences for death (\u03c7\u00b2\u2009=\u20099.30,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.158). In conclusion, LoRA fine-tuned LLM showed performance broadly comparable to statistical and machine learning models, with significant positive \u0394R mainly observed in case-forecasting and in zoonotic and intestinal infectious diseases.\n                  <\/jats:p>","DOI":"10.1371\/journal.pdig.0001527","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T17:41:59Z","timestamp":1783532519000},"page":"e0001527","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Large language models as versatile predictive engines for notifiable infectious diseases"],"prefix":"10.1371","volume":"5","author":[{"given":"Xinsheng","family":"Wu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinyuan","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhongwen","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyue","family":"Feng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ye","family":"Yao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jason J.","family":"Ong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huachun","family":"Zou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,8]]},"reference":[{"key":"pdig.0001527.ref001","article-title":"Precision public health must be at the centre of modern healthcare","volume":"378","author":"D Prior","year":"2022","journal-title":"BMJ"},{"key":"pdig.0001527.ref002","article-title":"How US law will evaluate artificial intelligence for covid-19","volume":"372","author":"M Krass","year":"2021","journal-title":"BMJ"},{"key":"pdig.0001527.ref003","doi-asserted-by":"crossref","first-page":"333","DOI":"10.1200\/JCO.2020.39.28_suppl.333","article-title":"Machine learning-based approach to the risk assessment of potentially preventable outpatient cancer treatment-related emergency care and hospitalizations","volume":"39","author":"K Miao","year":"2021","journal-title":"JCO"},{"issue":"10236","key":"pdig.0001527.ref004","doi-asserted-by":"crossref","first-page":"1579","DOI":"10.1016\/S0140-6736(20)30226-9","article-title":"Artificial intelligence and the future of global health","volume":"395","author":"N Schwalbe","year":"2020","journal-title":"Lancet"},{"key":"pdig.0001527.ref005","doi-asserted-by":"crossref","DOI":"10.2196\/68442","article-title":"Two-Year Hypertension Incidence Risk Prediction in Populations in the Desert Regions of Northwest China: Prospective Cohort Study","volume":"27","author":"Y Cheng","year":"2025","journal-title":"J Med Internet Res"},{"issue":"1","key":"pdig.0001527.ref006","article-title":"Evaluation of predictive performance of modeling hyperuricemia using medical big data: comparison of data preprocessing methods","volume":"12","author":"L Li","year":"2025","journal-title":"J Big Data"},{"key":"pdig.0001527.ref007","article-title":"Early Warning and Prediction of Scarlet Fever in China Using the Baidu Search Index and Autoregressive Integrated Moving Average With Explanatory Variable (ARIMAX) Model: Time Series Analysis","volume":"25","author":"T Luo","year":"2023","journal-title":"J Med Internet Res"},{"key":"pdig.0001527.ref008","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1038\/s41746-020-0256-0","article-title":"Machine learning prediction of incidence of Alzheimer\u2019s disease using large-scale administrative health data","volume":"3","author":"JH Park","year":"2020","journal-title":"NPJ Digit Med"},{"issue":"1","key":"pdig.0001527.ref009","doi-asserted-by":"crossref","first-page":"97","DOI":"10.1109\/TKDE.2013.109","article-title":"Data mining with big data","volume":"26","author":"Wu Xindong","year":"2014","journal-title":"IEEE Trans Knowl Data Eng"},{"issue":"2","key":"pdig.0001527.ref010","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1093\/nsr\/nwt032","article-title":"Challenges of Big Data Analysis","volume":"1","author":"J Fan","year":"2014","journal-title":"Natl Sci Rev"},{"issue":"1","key":"pdig.0001527.ref011","doi-asserted-by":"crossref","first-page":"780","DOI":"10.1186\/s12889-025-21982-3","article-title":"Study on the prediction performance of AIDS monthly incidence in Xinjiang based on time series and deep learning models","volume":"25","author":"D Tang","year":"2025","journal-title":"BMC Public Health"},{"issue":"1","key":"pdig.0001527.ref012","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1007\/s10916-024-02076-w","article-title":"A Novel Deep Learning Approach for Forecasting Myocardial Infarction Occurrences with Time Series Patient Data","volume":"48","author":"MS Sayed","year":"2024","journal-title":"J Med Syst"},{"issue":"12","key":"pdig.0001527.ref013","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2020-039676","article-title":"Comparison of ARIMA model and XGBoost model for prediction of human brucellosis in mainland China: a time-series study","volume":"10","author":"M Alim","year":"2020","journal-title":"BMJ Open"},{"issue":"6","key":"pdig.0001527.ref014","doi-asserted-by":"crossref","DOI":"10.2196\/24285","article-title":"Real-time Prediction of the Daily Incidence of COVID-19 in 215 Countries and Territories Using Machine Learning: Model Development and Validation","volume":"23","author":"Y Peng","year":"2021","journal-title":"J Med Internet Res"},{"issue":"3","key":"pdig.0001527.ref015","doi-asserted-by":"crossref","first-page":"2917","DOI":"10.1007\/s11227-023-05567-8","article-title":"Forecasting the future popularity of the anti-vax narrative on Twitter with machine learning","volume":"80","author":"I Biri","year":"2023","journal-title":"J Supercomput"},{"key":"pdig.0001527.ref016","doi-asserted-by":"crossref","first-page":"103145","DOI":"10.1016\/j.inffus.2025.103145","article-title":"LLM-DG: Leveraging large language model for enhanced disease prediction via inter-patient and intra-patient modeling","volume":"121","author":"Y Kang","year":"2025","journal-title":"Inform Fusion"},{"issue":"8","key":"pdig.0001527.ref017","doi-asserted-by":"crossref","first-page":"1930","DOI":"10.1038\/s41591-023-02448-8","article-title":"Large language models in medicine","volume":"29","author":"AJ Thirunavukarasu","year":"2023","journal-title":"Nat Med"},{"key":"pdig.0001527.ref018","unstructured":"Radford A, Narasimhan K. Improving language understanding by generative pre-training. 2018."},{"key":"pdig.0001527.ref019","article-title":"Language Models are Few-Shot Learners. ArXiv","author":"TB Brown","year":"2020"},{"key":"pdig.0001527.ref020","unstructured":"Bai S, Chen K, Liu X, et al. Qwen2.5-VL Technical Report. 2025."},{"key":"pdig.0001527.ref021","doi-asserted-by":"crossref","unstructured":"Delfani E, Pappas N. Optimizing information freshness in IoT systems with update rate constraints: A token-based approach. 2024.","DOI":"10.23919\/IFIPNetworking62109.2024.10619779"},{"key":"pdig.0001527.ref022","doi-asserted-by":"crossref","DOI":"10.2196\/57358","article-title":"Enhancing Patient Outcome Prediction Through Deep Learning With Sequential Diagnosis Codes From Structured Electronic Health Record Data: Systematic Review","volume":"27","author":"T Hama","year":"2025","journal-title":"J Med Internet Res"},{"issue":"40","key":"pdig.0001527.ref023","doi-asserted-by":"crossref","first-page":"4291","DOI":"10.1093\/eurheartj\/ehae465","article-title":"Artificial intelligence in cardiovascular medicine: clinical applications","volume":"45","author":"TF L\u00fcscher","year":"2024","journal-title":"Eur Heart J"},{"issue":"1","key":"pdig.0001527.ref024","article-title":"Artificial intelligence, ChatGPT, and other large language models for social determinants of health: Current state and future directions","volume":"5","author":"JCL Ong","year":"2024","journal-title":"Cell Rep Med"},{"issue":"10","key":"pdig.0001527.ref025","article-title":"The potential for large language models to transform cardiovascular medicine","volume":"6","author":"G Quer","year":"2024","journal-title":"Lancet Digit Health"},{"key":"pdig.0001527.ref026","article-title":"Large Language Models Are Zero-Shot Time Series Forecasters. ArXiv","author":"N Gruver","year":"2023"},{"key":"pdig.0001527.ref027","article-title":"Time-LLM: Time Series Forecasting by Reprogramming Large Language Models. ArXiv","author":"M Jin","year":"2023"},{"issue":"4","key":"pdig.0001527.ref028","doi-asserted-by":"crossref","first-page":"437","DOI":"10.1016\/S0169-2070(00)00065-0","article-title":"Out-of-sample tests of forecasting accuracy: an analysis and review","volume":"16","author":"LJ Tashman","year":"2000","journal-title":"Int J Forecast"},{"key":"pdig.0001527.ref029","unstructured":"Ansari AF, Stella L, Turkmen C, et al. Chronos: Learning the Language of Time Series. https:\/\/arxiv.org\/abs\/2403.07815"},{"key":"pdig.0001527.ref030","article-title":"TimeGPT-1","author":"A Garza","year":"2023"},{"key":"pdig.0001527.ref031","doi-asserted-by":"crossref","first-page":"109370","DOI":"10.1016\/j.compeleceng.2024.109370","article-title":"A review of Explainable Artificial Intelligence in healthcare","volume":"118","author":"Z Sadeghi","year":"2024","journal-title":"Comput Electr Eng"},{"key":"pdig.0001527.ref032","first-page":"276","volume-title":"What does BERT look at? An analysis of BERT\u2019s attention","author":"K Clark","year":"2019"},{"key":"pdig.0001527.ref033","doi-asserted-by":"crossref","unstructured":"Abnar S, Zuidema W. Quantifying attention flow in transformers. In: Association for Computational Linguistics. 2020. p. 4190\u20137.","DOI":"10.18653\/v1\/2020.acl-main.385"},{"issue":"6","key":"pdig.0001527.ref034","doi-asserted-by":"crossref","first-page":"3915","DOI":"10.1109\/COMST.2025.3526606","article-title":"A Survey on Intelligent Network Operations and Performance Optimization Based on Large Language Models","volume":"27","author":"S Long","year":"2025","journal-title":"IEEE Commun Surv Tutorials"},{"key":"pdig.0001527.ref035","article-title":"Reasoning Abilities of Large Language Models: In-Depth Analysis on the Abstraction and Reasoning Corpus","author":"S Lee","year":"2025","journal-title":"ACM Trans Intell Syst Technol"},{"issue":"12","key":"pdig.0001527.ref036","doi-asserted-by":"crossref","first-page":"2983","DOI":"10.1038\/s41591-023-02594-z","article-title":"Large language models should be used as scientific reasoning engines, not knowledge databases","volume":"29","author":"D Truhn","year":"2023","journal-title":"Nat Med"},{"issue":"3","key":"pdig.0001527.ref037","doi-asserted-by":"crossref","first-page":"315","DOI":"10.1007\/s41019-025-00285-y","article-title":"Large Language Model Enhanced Knowledge Representation Learning: A Survey","volume":"10","author":"X Wang","year":"2025","journal-title":"Data Sci Eng"},{"issue":"6","key":"pdig.0001527.ref038","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3516367","article-title":"Self-Supervised Transformer for Sparse and Irregularly Sampled Multivariate Clinical Time-Series","volume":"16","author":"S Tipirneni","year":"2022","journal-title":"ACM Trans Knowl Discov Data"},{"key":"pdig.0001527.ref039","article-title":"Reframing demand forecasting: a two-fold approach for lumpy and interm ittent demand. arxiv:210313812[csLG].","author":"JM Ro\u017eanec"},{"key":"pdig.0001527.ref040","doi-asserted-by":"crossref","first-page":"100883","DOI":"10.1016\/j.cosrev.2025.100883","article-title":"A survey of transformer networks for time series forecasting","volume":"60","author":"J Zhao","year":"2026","journal-title":"Comput Sci Rev"},{"issue":"10","key":"pdig.0001527.ref041","doi-asserted-by":"crossref","first-page":"7790","DOI":"10.1109\/JBHI.2025.3561214","article-title":"MIFlu: Large Language Model-Based Multimodal Influenza Forecasting Scheme","volume":"29","author":"J Moon","year":"2025","journal-title":"IEEE J Biomed Health Inform"},{"issue":"9","key":"pdig.0001527.ref042","doi-asserted-by":"crossref","first-page":"992","DOI":"10.1016\/j.ajic.2024.03.016","article-title":"Utilizing natural language processing and large language models in the diagnosis and prediction of infectious diseases: A systematic review","volume":"52","author":"M Omar","year":"2024","journal-title":"Am J Infect Control"},{"key":"pdig.0001527.ref043","doi-asserted-by":"crossref","first-page":"3254","DOI":"10.1016\/j.csbj.2024.08.006","article-title":"Utilizing large language models in infectious disease transmission modelling for public health preparedness","volume":"23","author":"KO Kwok","year":"2024","journal-title":"Comput Struct Biotechnol J"},{"key":"pdig.0001527.ref044","doi-asserted-by":"crossref","unstructured":"Ruga T, Vocaturo E, Zumpano E. On the Role of LLM to Forecast the Next Pandemic. In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). 2024. p. 6567\u201373.","DOI":"10.1109\/BIBM62325.2024.10822835"},{"issue":"349","key":"pdig.0001527.ref045","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1080\/01621459.1975.10480264","article-title":"Intervention Analysis with Applications to Economic and Environmental Problems","volume":"70","author":"GEP Box","year":"1975","journal-title":"J Am Stat Assoc"},{"issue":"8","key":"pdig.0001527.ref046","doi-asserted-by":"crossref","first-page":"685","DOI":"10.1016\/S2213-2600(23)00050-4","article-title":"Predicting the future risk of lung cancer: development, and internal and external validation of the CanPredict (lung) model in 19\u00b767 million people and evaluation of model performance against seven other risk prediction models","volume":"11","author":"W Liao","year":"2023","journal-title":"Lancet Respir Med"},{"key":"pdig.0001527.ref047","article-title":"Predicting recurrence of venous thromboembolism in anticoagulated cancer patients using real-world data and machine learning","volume":"40","author":"AJ Mu\u00f1oz Mart\u00edn","year":"2022","journal-title":"JCO"},{"key":"pdig.0001527.ref048","article-title":"Predicting malaria outbreaks using earth observation measurements and spatiotemporal deep learning modelling: a South Asian case study from 2000 to 2017","author":"U Nazir","year":"2024","journal-title":"Lancet Planet Health"},{"key":"pdig.0001527.ref049","doi-asserted-by":"crossref","DOI":"10.2196\/66072","article-title":"Forecasting the Incidence of Mumps Based on the Baidu Index and Environmental Data in Yunnan, China: Deep Learning Model Study","volume":"27","author":"X Xiong","year":"2025","journal-title":"J Med Internet Res"},{"issue":"1","key":"pdig.0001527.ref050","doi-asserted-by":"crossref","first-page":"7913","DOI":"10.1038\/s41467-023-43713-1","article-title":"Augmenting interpretable models with large language models during training","volume":"14","author":"C Singh","year":"2023","journal-title":"Nat Commun"},{"issue":"2","key":"pdig.0001527.ref051","doi-asserted-by":"crossref","first-page":"317","DOI":"10.1016\/j.ejor.2023.09.010","article-title":"Explainability in process outcome prediction: Guidelines to obtain interpretable and faithful models","volume":"317","author":"A Stevens","year":"2024","journal-title":"Eur J Operat Res"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001527","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T17:42:10Z","timestamp":1783532530000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001527"}},"subtitle":[],"editor":[{"given":"Dhiya","family":"Al-Jumeily OBE","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,8]]},"references-count":51,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,8]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001527","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,8]]}}}