{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T23:16:20Z","timestamp":1793142980927,"version":"4.5.0"},"reference-count":42,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T00:00:00Z","timestamp":1785110400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T00:00:00Z","timestamp":1785110400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Zewail City of Science & Technology"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Sepsis is one of the most deadly illnesses with a high risk of mortality. Consequently, identifying it at the beginning of illness symptoms is crucial and plays a key role in improving patient outcomes. This study presents a customized solution for the early detection of sepsis with an emphasis on the use of interpretability and explainability techniques, utilizing a range of machine learning approaches and interpretable artificial intelligence methods. The database on which this research study is based has many problems; the main ones being large data gaps and class disparities. Employing robust methods, precise categorizations, and rigorous computations, Approximately 12 diverse models were developed and optimized. With ROC-AUC indicators of 0.9566 and 0.9595 and F1 scores of 0.85 and 0.85 respectively, Bidirectional Long Short-Term Memory (BiLSTM) and Temporal Convolutional Network (TCN) models performed better than conventional models in terms of sepsis prediction. These two approaches have shown remarkable progress in detecting clinical patterns while avoiding false negative results\u2014an essential aspect of the medical field. To assess model performance and offer clear insights into model predictions, interpretation-based techniques were employed. This improved clinical confidence and facilitated well-informed decisions in crucial medical diagnoses.<\/jats:p>","DOI":"10.1038\/s41598-026-61652-x","type":"journal-article","created":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T09:02:09Z","timestamp":1785142929000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Explainable deep learning for early sepsis detection from ICU time-series data using XAI techniques"],"prefix":"10.1038","volume":"16","author":[{"given":"Anas","family":"Mahmoud","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hamza","family":"Abdelmoreed","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hossam","family":"Amir","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohamed","family":"Ehab","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rana","family":"Abdelfattah","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mayada","family":"HadHoud","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,27]]},"reference":[{"key":"61652_CR1","unstructured":"World Health Organization. WHO calls for global action on sepsis\u2014cause of 1 in 5 deaths worldwide. 8, 2020. [Online]. Available: https:\/\/www.who.int\/news\/item\/08-09-2020-who-calls-for-global-action-on-sepsis---cause-of-1-in-5-deaths-worldwide"},{"issue":"7","key":"61652_CR2","first-page":"409","volume":"101","author":"RL Gauer","year":"2020","unstructured":"Gauer, R. L., Braun, M. L. & Ott, M. C. Sepsis: diagnosis and management. Am. Fam. Physician 101(7), 409\u2013418 (2020).","journal-title":"Am. Fam. Physician"},{"issue":"2","key":"61652_CR3","doi-asserted-by":"publisher","first-page":"210","DOI":"10.1097\/CCM.0000000000004145","volume":"48","author":"MA Reyna","year":"2020","unstructured":"Reyna, M. A. et al. Early prediction of sepsis from clinical data: The PhysioNet\/Computing in Cardiology Challenge 2019. Crit. Care Med. 48(2), 210\u2013217. https:\/\/doi.org\/10.1097\/CCM.0000000000004145 (2020).","journal-title":"Crit. Care Med."},{"issue":"6","key":"61652_CR4","doi-asserted-by":"publisher","first-page":"1589","DOI":"10.1097\/01.CCM.0000217961.75225.EF","volume":"34","author":"A Kumar","year":"2006","unstructured":"Kumar, A. et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit. Care Med. 34(6), 1589\u20131596. https:\/\/doi.org\/10.1097\/01.CCM.0000217961.75225.EF (2006).","journal-title":"Crit. Care Med."},{"issue":"12","key":"61652_CR5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s11908-021-00765-y","volume":"23","author":"TW van der Vaart","year":"2021","unstructured":"van der Vaart, T. W., Huguenard, J. F. & Rogers, A. J. Diagnostic challenges in sepsis. Curr. Infect. Dis. Rep. 23(12), 1\u20139. https:\/\/doi.org\/10.1007\/s11908-021-00765-y (2021).","journal-title":"Curr. Infect. Dis. Rep."},{"key":"61652_CR6","doi-asserted-by":"publisher","unstructured":"Lyra, S., Leonhardt, S. & Antink, C.H. Early prediction of sepsis using random forest classification for imbalanced clinical data. Comput. Cardiol.. https:\/\/doi.org\/10.22489\/CinC.2019.071 (2019).","DOI":"10.22489\/CinC.2019.071"},{"key":"61652_CR7","doi-asserted-by":"publisher","unstructured":"Wu, Q., Li, Y., Liu, X. et al. A customized down-sampling machine learning approach for sepsis prediction. Comput. Cardiol.. https:\/\/doi.org\/10.22489\/CinC.2019.071 (2019).","DOI":"10.22489\/CinC.2019.071"},{"key":"61652_CR8","unstructured":"Liao, J., Li, Y., Liu, X. et al. Does deep learning REALLY outperform non-deep machine learning for clinical prediction on physiological time series? arXiv preprint, arXiv:2211.06034, (2022)."},{"key":"61652_CR9","doi-asserted-by":"publisher","unstructured":"Nirgudkar, S. & Ding, T. Early detection of sepsis using ensembles. Comput. Cardiol.. https:\/\/doi.org\/10.22489\/CinC.2019.071 (2019).","DOI":"10.22489\/CinC.2019.071"},{"key":"61652_CR10","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2019.103940","volume":"131","author":"G Kong","year":"2019","unstructured":"Kong, G., Lin, X. & Hu, H. Using machine learning methods to predict sepsis. Int. J. Med. Inform. 131, 103940. https:\/\/doi.org\/10.1016\/j.ijmedinf.2019.103940 (2019).","journal-title":"Int. J. Med. Inform."},{"key":"61652_CR11","doi-asserted-by":"crossref","unstructured":"Vandana S. & Chhikara, R. Comparative analysis of ML models with selection methods for early predictive analytics of sepsis in ICU Indian. J. Sci. Technol. 17(41), 4338\u20134348. https:\/\/doi.org\/10.17485\/IJST\/v17i41.1925 (2024).","DOI":"10.17485\/IJST\/v17i41.1925"},{"key":"61652_CR12","unstructured":"Lipton, Z.C., Kale, D.C., Elkan, C. & Wetzel, R. Modeling missing data in clinical time series with RNNs. arXiv preprint. arXiv:1606.04130 (2016). [Online]."},{"key":"61652_CR13","doi-asserted-by":"crossref","unstructured":"Liu, L., Li, Y., Liu, X. et al. Early prediction of sepsis from clinical data via heterogeneous event aggregation. arXiv preprint. arXiv:1910.06792 (2019). [Online].","DOI":"10.22489\/CinC.2019.157"},{"key":"61652_CR14","doi-asserted-by":"publisher","first-page":"132437","DOI":"10.1109\/ACCESS.2020.3010445","volume":"8","author":"K Apalak","year":"2020","unstructured":"Apalak, K. & Kiasaleh, K. Early detection of sepsis with temporal convolutional networks. IEEE Access 8, 132437\u2013132449. https:\/\/doi.org\/10.1109\/ACCESS.2020.3010445 (2020).","journal-title":"IEEE Access"},{"key":"61652_CR15","doi-asserted-by":"publisher","DOI":"10.1016\/j.ibmed.2021.100042","volume":"5","author":"SP Oei","year":"2021","unstructured":"Oei, S. P. et al. Towards early sepsis detection from measurements at the general ward through deep learning. Intelligence-Based Med. 5, 100042. https:\/\/doi.org\/10.1016\/j.ibmed.2021.100042 (2021).","journal-title":"Intelligence-Based Med."},{"issue":"1","key":"61652_CR16","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1002\/mpr.329","volume":"20","author":"MJ Azur","year":"2011","unstructured":"Azur, M. J., Stuart, E. A., Frangakis, C. & Leaf, P. J. Multiple imputation by chained equations: What is it and how does it work? Int. J. Methods Psychiatr. Res. 20(1), 40\u201349. https:\/\/doi.org\/10.1002\/mpr.329 (2011).","journal-title":"Int. J. Methods Psychiatr. Res."},{"key":"61652_CR17","unstructured":"Hong, Y., Zhou, J., Ji, S. & Fang, Y. Sepsis prediction using InceptionTime architecture. In Proceedings of the 2021 International Conference on Smart City and Ubiquitous Computing (2021)."},{"key":"61652_CR18","unstructured":"Mohammad, F., Li, Y., Liu, X. et al. Early prediction of onset of sepsis in a clinical setting. arXiv preprint. arXiv:2402.03486 (2024)."},{"key":"61652_CR19","doi-asserted-by":"publisher","unstructured":"Chen, T., Guestrin, C. XGBoost: a scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 785\u2013794. https:\/\/doi.org\/10.1145\/2939672.2939785 (New York, NY, USA, 2016).","DOI":"10.1145\/2939672.2939785"},{"key":"61652_CR20","first-page":"2546","volume":"24","author":"J Bergstra","year":"2011","unstructured":"Bergstra, J., Bardenet, R., Bengio, Y. & K\u00e9gl, B. Algorithms for hyper-parameter optimization. Adv. Neural Inf. Process. Syst. 24, 2546\u20132554 (2011).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"61652_CR21","doi-asserted-by":"publisher","unstructured":"Yang, M., Wang, X., Gao, H., Li, Y., Liu, X., Li, J. & Liu, C. Early prediction of sepsis using Multi-Feature fusion based XGBoost learning and Bayesian optimization. Comput. Cardiol.. https:\/\/doi.org\/10.22489\/CinC.2019.020 (2019).","DOI":"10.22489\/CinC.2019.020"},{"issue":"1","key":"61652_CR22","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1038\/s41597-019-0103-9","volume":"6","author":"H Harutyunyan","year":"2019","unstructured":"Harutyunyan, H., Khachatrian, H., Kale, D. C. & Galstyan, A. Multitask learning and benchmarking with clinical time series data. Sci. Data 6(1), 96. https:\/\/doi.org\/10.1038\/s41597-019-0103-9 (2019).","journal-title":"Sci. Data"},{"issue":"1","key":"61652_CR23","doi-asserted-by":"publisher","first-page":"6313","DOI":"10.1038\/s41598-021-85700-w","volume":"11","author":"N Takahashi","year":"2021","unstructured":"Takahashi, N., Nakada, T.-A., Walley, K. R. & Russell, J. A. Significance of lactate clearance in septic shock patients with high bilirubin levels. Sci. Rep. 11(1), 6313. https:\/\/doi.org\/10.1038\/s41598-021-85700-w (2021).","journal-title":"Sci. Rep."},{"key":"61652_CR24","unstructured":"Klabunde, R. E. Cardiovascular Physiology Concepts 2nd edn. 9781451113846 (Lippincott Williams & Wilkins, Philadelphia, PA, 2012)."},{"key":"61652_CR25","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35","volume":"3","author":"AE Johnson","year":"2016","unstructured":"Johnson, A. E. et al. MIMIC-III, a freely accessible critical care database. Sci. Data 3, 160035. https:\/\/doi.org\/10.1038\/sdata.2016.35 (2016).","journal-title":"Sci. Data"},{"key":"61652_CR26","doi-asserted-by":"publisher","first-page":"148294","DOI":"10.1109\/ACCESS.2024.3470851","volume":"12","author":"S Lee","year":"2024","unstructured":"Lee, S. et al. Early prediction of sepsis in the intensive care unit using the GRU-D-MGP-TCN model. IEEE Access 12, 148294\u2013148304. https:\/\/doi.org\/10.1109\/ACCESS.2024.3470851 (2024).","journal-title":"IEEE Access"},{"key":"61652_CR27","doi-asserted-by":"publisher","first-page":"3417","DOI":"10.1109\/ACCESS.2023.3449242","volume":"12","author":"M Apalak","year":"2024","unstructured":"Apalak, M. & Kiasaleh, K. Advancing early detection of sepsis with temporal convolutional networks using ECG signals. IEEE Access 12, 3417\u20133427. https:\/\/doi.org\/10.1109\/ACCESS.2023.3449242 (2024).","journal-title":"IEEE Access"},{"key":"61652_CR28","unstructured":"Venkatesh, B., Pradeep, P.K.R., Ramadevi, Y., Vishwa Kiran, V.K.S. & Uday, K. Leveraging deep learning for real-time, continuous monitoring and prediction of sepsis in ICU patients using multisensory data. J. Theoretical Appl. Inf.Technol. 103(10), 4115\u20134128 (2025)."},{"key":"61652_CR29","unstructured":"Futoma, J., Hariharan, S., Heller, K., Sendak, M., Brajer, N., Clement, M., Bedoya, A. & O\u2019Brien, C. An improved multi-output gaussian process RNN with real-time validation for early sepsis detection. arXiv preprint arXiv:1708.05894 (2017)."},{"issue":"8","key":"61652_CR30","doi-asserted-by":"publisher","first-page":"871","DOI":"10.2169\/internalmedicine.51.6951","volume":"51","author":"T Kenzaka","year":"2012","unstructured":"Kenzaka, T. et al. Importance of vital signs to the early diagnosis and severity of sepsis: association between vital signs and sequential organ failure assessment score in patients with sepsis. Intern. Med. 51(8), 871\u2013876. https:\/\/doi.org\/10.2169\/internalmedicine.51.6951 (2012).","journal-title":"Intern. Med."},{"key":"61652_CR31","unstructured":"Alliance, S. Symptoms - Sepsis Alliance, Available: https:\/\/www.sepsis.org\/sepsis-basics\/symptoms\/ (2023)."},{"issue":"10","key":"61652_CR32","doi-asserted-by":"publisher","first-page":"1596","DOI":"10.1093\/jamia\/ocac093","volume":"29","author":"ES Rangan","year":"2022","unstructured":"Rangan, E. S. et al. Performance effectiveness of vital parameter combinations for early warning of sepsis. J. Am. Med. Inform. Assoc. 29(10), 1596\u20131606. https:\/\/doi.org\/10.1093\/jamia\/ocac093 (2022).","journal-title":"J. Am. Med. Inform. Assoc."},{"issue":"4","key":"61652_CR33","doi-asserted-by":"publisher","first-page":"329","DOI":"10.1007\/s10877-013-9530-x","volume":"28","author":"KD Fairchild","year":"2014","unstructured":"Fairchild, K. D. et al. Complex signals bioinformatics: Evaluation of heart rate characteristics monitoring as a novel risk marker for neonatal sepsis. J. Clin. Monit. Comput. 28(4), 329\u2013339. https:\/\/doi.org\/10.1007\/s10877-013-9530-x (2014).","journal-title":"J. Clin. Monit. Comput."},{"key":"61652_CR34","doi-asserted-by":"crossref","unstructured":"Kausch, S.L., Rogers, A.J., et al. A cardiorespiratory signature of sepsis in 3 NICUs, medRxiv. https:\/\/doi.org\/10.1101\/2022.09.28.22280469 (2022).","DOI":"10.1101\/2022.09.28.22280469"},{"key":"61652_CR35","unstructured":"Infectious Disease Advisor. Systolic Blood Pressure Trajectory May Influence In-Hospital Mortality Risk in Sepsis. https:\/\/www.infectiousdiseaseadvisor.com\/news\/systolic-blood-pressure-trajectory-may-influence-in-hospital-mortality-risk-in-sepsis\/ (2024)."},{"issue":"1","key":"61652_CR36","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1186\/s12879-023-08054-w","volume":"23","author":"JL Zhu","year":"2023","unstructured":"Zhu, J. L. et al. Influence of systolic blood pressure trajectory on in-hospital mortality in patients with sepsis. BMC Infect. Dis. 23(1), 90. https:\/\/doi.org\/10.1186\/s12879-023-08054-w (2023).","journal-title":"BMC Infect. Dis."},{"issue":"1","key":"61652_CR37","doi-asserted-by":"publisher","first-page":"97","DOI":"10.1186\/s12873-023-00864-0","volume":"23","author":"SY Liu","year":"2023","unstructured":"Liu, S. Y. et al. Association between diastolic blood pressure during the first 24 h and 28-day mortality in patients with septic shock: A retrospective observational study. BMC Emerg. Med. 23(1), 97. https:\/\/doi.org\/10.1186\/s12873-023-00864-0 (2023).","journal-title":"BMC Emerg. Med."},{"key":"61652_CR38","doi-asserted-by":"publisher","unstructured":"Ranjit, S. & Natraj, R. Hemodynamic assessment and management of septic shock in children. J. Pediatr. Crit. Care 11(10), 186\u2013194. https:\/\/doi.org\/10.5005\/jp-journals-10071-24246 (2024).","DOI":"10.5005\/jp-journals-10071-24246"},{"key":"61652_CR39","doi-asserted-by":"publisher","unstructured":"Berger, T. et al. Shock Index and Early Recognition of Sepsis in the Emergency Department: Pilot Study. West J. Emerg. Med. 14(2), 168\u2013174. https:\/\/doi.org\/10.5811\/westjem.2012.8.11546 (2013).","DOI":"10.5811\/westjem.2012.8.11546"},{"key":"61652_CR40","doi-asserted-by":"crossref","unstructured":"Wachter, S., Mittelstadt, B. & Russell, C. Counterfactual explanations without opening the black box: Automated decisions and the GDPR. Harvard J. Law Technol. 31(2), 841\u2013887. arxiv: 1711.00399 (2017).","DOI":"10.2139\/ssrn.3063289"},{"key":"61652_CR41","unstructured":"Kim, B., Wattenberg, M., Gilmer, J., Cai, C., Wexler, J., Viegas, F. & Sayres, R. Interpretability beyond feature attribution: quantitative testing with concept activation vectors (TCAV). In Proc. 35th Int. Conf. Machine Learning (ICML), 2668\u20132677. arxiv: 1711.11279 (2018)."},{"key":"61652_CR42","doi-asserted-by":"publisher","DOI":"10.1038\/sdata201635","volume":"3","author":"AEW Johnson","year":"2016","unstructured":"Johnson, A. E. W. et al. MIMIC-III, a freely accessible critical care database. Sci. Data 3, 160035. https:\/\/doi.org\/10.1038\/sdata201635 (2016).","journal-title":"Scientific Data"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-61652-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-61652-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-61652-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T09:02:13Z","timestamp":1785142933000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-61652-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,27]]},"references-count":42,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["61652"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-61652-x","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,27]]},"assertion":[{"value":"8 October 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 July 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study utilized the publicly available PhysioNet 2019 Challenge dataset, which has been previously de-identified and approved for research use. No additional ethics approval was required for this retrospective analysis of anonymized clinical data.","order":2,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable. This study used publicly available anonymized data.","order":3,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"23217"}}