{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T22:47:09Z","timestamp":1789512429173,"version":"build-2803163510"},"reference-count":34,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T00:00:00Z","timestamp":1778025600000},"content-version":"unspecified","delay-in-days":5,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100017510","name":"UK Dementia Research Institute","doi-asserted-by":"publisher","award":["UK DRI-7002"],"award-info":[{"award-number":["UK DRI-7002"]}],"id":[{"id":"10.13039\/501100017510","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Health Care Inform"],"accepted":{"date-parts":[[2026,3,25]]},"published-print":{"date-parts":[[2026,5]]},"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Sepsis is a potentially fatal systemic response to infection, in which early clinical intervention is critical to reduce mortality. This study presents a hybrid deep learning model that combines temporal and structural information from clinical data to improve early sepsis prediction.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We used data from the PhysioNet\/Computing in Cardiology Challenge 2019 to predict sepsis onset up to 12 hours in advance. We developed a hybrid model integrating long short-term memory (LSTM) networks and graph attention networks (GAT) to capture temporal dynamics and intervariable relationships. Performance was compared with three baseline models. To ensure robustness, all models were trained using five repeated train-test splits with different random seeds.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The dataset includes 40\u2009336 adult ICU patients. Of all the patients, 2932 developed sepsis during their stay. Each patient\u2019s data includes hourly data on 40 clinical variables, including vital signs, laboratory results and demographic information. The LSTM-GAT model achieved an area under the receiver operating characteristic curve (AUROC) of 0.853\u00b10.005, an F1-score of 0.627\u00b10.006 and a specificity of 0.872\u00b10.007, outperforming baseline models. Despite being trained on fixed temporal windows, the model generalised well across multiple prediction horizons without retraining.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>By integrating temporal and structural representations, the proposed approach achieves improved predictive performance compared with the baseline. This capability may support earlier identification of high-risk patients and enhance timely clinical decision-making in critical care environments.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed model demonstrates the advantage of combining sequence and graph-based methods. It offers a promising tool for real-time clinical decision support in sepsis detection.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjhci-2025-101762","type":"journal-article","created":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T15:45:19Z","timestamp":1778082319000},"page":"e101762","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":1,"title":["Early sepsis prediction using a hybrid LSTM-GAT model: a study on the PhysioNet 2019 dataset"],"prefix":"10.1136","volume":"33","author":[{"given":"Bahar","family":"Khorram","sequence":"first","affiliation":[{"name":"Department of Electrical and Electronic Engineering, University of Surrey, Guildford, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Samaneh","family":"Kouchaki","sequence":"additional","affiliation":[{"name":"Department of Electrical and Electronic Engineering, University of Surrey, Guildford, UK"},{"name":"Care Research and Technology Centre, UK Dementia Research Institute, London, England, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,5,6]]},"reference":[{"key":"2026050608451147000_33.1.e101762.1","unstructured":"World Health Organization . Sepsis. 2020. Available: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/sepsis"},{"key":"2026050608451147000_33.1.e101762.2","doi-asserted-by":"crossref","first-page":"200","DOI":"10.1016\/S0140-6736(19)32989-7","article-title":"Global, regional, and national sepsis incidence and mortality, 1990\u20132017: analysis for the Global Burden of Disease Study","volume":"395","author":"Rudd","year":"2020","journal-title":"The Lancet"},{"key":"2026050608451147000_33.1.e101762.3","doi-asserted-by":"crossref","DOI":"10.1186\/s13054-020-02950-2","article-title":"Mortality in sepsis and septic shock in Europe, North America and Australia between 2009 and 2019- results from a systematic review and meta-analysis","volume":"24","author":"Bauer","year":"2020","journal-title":"Crit Care"},{"key":"2026050608451147000_33.1.e101762.4","doi-asserted-by":"crossref","first-page":"154096","DOI":"10.1016\/j.jcrc.2022.154096","article-title":"Hospital-related costs of sepsis around the world: A systematic review exploring the economic burden of sepsis","volume":"71","author":"van den Berg","year":"2022","journal-title":"J Crit Care"},{"key":"2026050608451147000_33.1.e101762.5","doi-asserted-by":"publisher","DOI":"10.1097\/CCM.0000000000004777"},{"key":"2026050608451147000_33.1.e101762.6","doi-asserted-by":"crossref","unstructured":"Haas R , McGill SC . Artificial Intelligence for the Prediction of Sepsis in Adults. cjht 2022;2. doi:10.51731\/cjht.2022.283","DOI":"10.51731\/cjht.2022.283"},{"key":"2026050608451147000_33.1.e101762.7","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-021-20910-4","article-title":"Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare","volume":"12","author":"Goh","year":"2021","journal-title":"Nat Commun"},{"key":"2026050608451147000_33.1.e101762.8","doi-asserted-by":"crossref","DOI":"10.1186\/s13054-024-04948-6","article-title":"A scoping review of machine learning for sepsis prediction- feature engineering strategies and model performance: a step towards explainability","volume":"28","author":"Bomrah","year":"2024","journal-title":"Crit Care"},{"key":"2026050608451147000_33.1.e101762.9","doi-asserted-by":"crossref","DOI":"10.3389\/fmed.2024.1491358","article-title":"Early detection of sepsis using machine learning algorithms: a systematic review and network meta-analysis","volume":"11","author":"Yadgarov","year":"2024","journal-title":"Front Med (Lausanne)"},{"key":"2026050608451147000_33.1.e101762.10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.cmpb.2018.12.027","article-title":"Prediction of sepsis patients using machine learning approach: A meta-analysis","volume":"170","author":"Islam","year":"2019","journal-title":"Comput Methods Programs Biomed"},{"key":"2026050608451147000_33.1.e101762.11","doi-asserted-by":"publisher","DOI":"10.1007\/s00134-019-05872-y"},{"key":"2026050608451147000_33.1.e101762.12","doi-asserted-by":"crossref","unstructured":"Reyna M , Josef C , Jeter R , et al . Early prediction of sepsis from clinical data: the physionet\/computing in cardiology challenge 2019 (version 1.0.0). PhysioNet 2019. doi:10.13026\/v64v-d857","DOI":"10.22489\/CinC.2019.412"},{"key":"2026050608451147000_33.1.e101762.13","doi-asserted-by":"crossref","unstructured":"Morrill J , Kormilitzin A , Nevado-Holgado A , et al . The signature-based model for early detection of sepsis from electronic health records in the intensive care unit. 2019 Computing in Cardiology Conference; IEEE, 1. doi:10.22489\/CinC.2019.014","DOI":"10.22489\/CinC.2019.014"},{"key":"2026050608451147000_33.1.e101762.14","unstructured":"Du JA , Sadr N , Chazal P . Automated prediction of sepsis onset using gradient boosted decision trees. 2019 Computing in Cardiology (CinC); 2019:1."},{"key":"2026050608451147000_33.1.e101762.15","doi-asserted-by":"crossref","unstructured":"Zabihi M , Kiranyaz S , Gabbouj M . Sepsis prediction in intensive care unit using ensemble of xgboost models. 2019 Computing in Cardiology Conference; 2019:1. doi:10.22489\/CinC.2019.238","DOI":"10.22489\/CinC.2019.238"},{"key":"2026050608451147000_33.1.e101762.16","doi-asserted-by":"crossref","DOI":"10.1155\/2019\/5930379","article-title":"Machine Learning Models for Analysis of Vital Signs Dynamics: A Case for Sepsis Onset Prediction","volume":"2019","author":"Bloch","year":"2019","journal-title":"J Healthc Eng"},{"key":"2026050608451147000_33.1.e101762.17","doi-asserted-by":"publisher","DOI":"10.1097\/CCM.0000000000002936"},{"key":"2026050608451147000_33.1.e101762.18","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-025-85121-z","article-title":"Interpretable machine learning for predicting sepsis risk in emergency triage patients","volume":"15","author":"Liu","year":"2025","journal-title":"Sci Rep"},{"key":"2026050608451147000_33.1.e101762.19","doi-asserted-by":"crossref","first-page":"101820","DOI":"10.1016\/j.artmed.2020.101820","article-title":"Early detection of sepsis utilizing deep learning on electronic health record event sequences","volume":"104","author":"Lauritsen","year":"2020","journal-title":"Artif Intell Med"},{"key":"2026050608451147000_33.1.e101762.20","doi-asserted-by":"crossref","first-page":"104110","DOI":"10.1016\/j.compbiomed.2020.104110","article-title":"SSP: Early prediction of sepsis using fully connected LSTM-CNN model","volume":"128","author":"Rafiei","year":"2021","journal-title":"Comput Biol Med"},{"key":"2026050608451147000_33.1.e101762.21","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2019.103395"},{"key":"2026050608451147000_33.1.e101762.22","doi-asserted-by":"crossref","first-page":"7156","DOI":"10.3390\/jcm12227156","article-title":"Development and Validation of Deep-Learning-Based Sepsis and Septic Shock Early Prediction System (DeepSEPS) Using Real-World ICU Data","volume":"12","author":"Kim","year":"2023","journal-title":"JCM"},{"key":"2026050608451147000_33.1.e101762.23","doi-asserted-by":"crossref","unstructured":"Wickramaratne SD , Shaad Mahmud M . Bi-directional gated recurrent unit based ensemble model for the early detection of sepsis. 2020 42nd Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC); IEEE, 70\u20133. Montreal, QC, Canada. doi:10.1109\/EMBC44109.2020.9175223","DOI":"10.1109\/EMBC44109.2020.9175223"},{"key":"2026050608451147000_33.1.e101762.24","unstructured":"Futoma J , Hariharan S , Heller K , et al . An improved multi-output gaussian process rnn with real-time validation for early sepsis detection. Machine Learning for Healthcare Conference; 2017:243\u201354."},{"key":"2026050608451147000_33.1.e101762.25","doi-asserted-by":"crossref","DOI":"10.1016\/j.eclinm.2023.102124","article-title":"Predicting sepsis using deep learning across international sites: a retrospective development and validation study","volume":"62","author":"Moor","year":"2023","journal-title":"EClinicalMedicine"},{"key":"2026050608451147000_33.1.e101762.26","unstructured":"Shashikumar SP , Josef C , Sharma A , et al . DeepAISE--An end-to-end development and deployment of a recurrent neural survival model for early prediction of sepsis. arXiv 2019."},{"key":"2026050608451147000_33.1.e101762.27","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-023-00986-6","article-title":"Impact of a deep learning sepsis prediction model on quality of care and survival","volume":"7","author":"Boussina","year":"2024","journal-title":"NPJ Digit Med"},{"key":"2026050608451147000_33.1.e101762.28","doi-asserted-by":"crossref","first-page":"e1106","DOI":"10.1097\/CCM.0000000000004583","article-title":"Graph Convolutional Networks-Based Noisy Data Imputation in Electronic Health Record","volume":"48","author":"Lee","year":"2020","journal-title":"Crit Care Med"},{"key":"2026050608451147000_33.1.e101762.29","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"2026050608451147000_33.1.e101762.30","unstructured":"Veli\u010dkovi\u0107 P , Cucurull G , Casanova A , et al . Graph attention networks. arXiv 2017."},{"key":"2026050608451147000_33.1.e101762.31","unstructured":"Johnson A , Pollard T , Mark R . Data from: MIMIC-III Clinical Database (version 1.4). PhysioNet 2016. doi:10.13026\/C2XW26"},{"key":"2026050608451147000_33.1.e101762.32","doi-asserted-by":"crossref","DOI":"10.3390\/ijerph19042349","article-title":"Early Prediction of Sepsis Onset Using Neural Architecture Search Based on Genetic Algorithms","volume":"19","author":"Kim","year":"2022","journal-title":"Int J Environ Res Public Health"},{"key":"2026050608451147000_33.1.e101762.33","doi-asserted-by":"crossref","first-page":"e1091","DOI":"10.1097\/CCM.0000000000004550","article-title":"An Explainable Artificial Intelligence Predictor for Early Detection of Sepsis","volume":"48","author":"Yang","year":"2020","journal-title":"Crit Care Med"},{"key":"2026050608451147000_33.1.e101762.34","doi-asserted-by":"crossref","unstructured":"Rangan ES , Pathinarupothi RK , Anand KJS , et al . Performance effectiveness of vital parameter combinations for early warning of sepsis-an exhaustive study using machine learning. JAMIA Open 2022;5. doi:10.1093\/jamiaopen\/ooac080","DOI":"10.1093\/jamiaopen\/ooac080"}],"container-title":["BMJ Health &amp; Care Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjhci-2025-101762","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T15:45:42Z","timestamp":1778082342000},"score":1,"resource":{"primary":{"URL":"https:\/\/informatics.bmj.com\/lookup\/doi\/10.1136\/bmjhci-2025-101762"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":34,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,5,6]]},"published-print":{"date-parts":[[2026,5]]}},"alternative-id":["10.1136\/bmjhci-2025-101762"],"URL":"https:\/\/doi.org\/10.1136\/bmjhci-2025-101762","relation":{},"ISSN":["2632-1009"],"issn-type":[{"value":"2632-1009","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]}}}