{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T01:15:53Z","timestamp":1793495753268,"version":"4.5.0"},"reference-count":34,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T00:00:00Z","timestamp":1778198400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:00:00Z","timestamp":1784073600000},"content-version":"vor","delay-in-days":68,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Acute kidney injury (AKI) is a common hospital complication with substantial morbidity and mortality. Deep learning models for AKI prediction show strong development-cohort performance, but single-point evaluation fails to capture behaviour under continuous monitoring. We conducted a multi-centre retrospective study using electronic health records from three cohorts (\n                    <jats:italic>n<\/jats:italic>\n                    \u2009=\u2009157,323 admissions): National Health Insurance Service Ilsan Hospital (development), Chuncheon Sacred Heart Hospital, and MIMIC-IV (external validation). Three deep learning architectures (LSTM-Attention, Masked CNN, ITE-Transformer) and two baselines (XGBoost, logistic regression) were developed across 0-, 48-, and 72-h horizons, with an online simulation framework generating predictions at 12-h intervals before onset. Deep learning substantially outperformed baselines externally (AUROC 0.956\u20130.963 vs. 0.630\u20130.686). Online simulation revealed that 0-h models exhibited \u201cclinical faithfulness\u201d\u2014consistent AUROC improvement as onset approached (Mann\u2013Kendall significant in 15\/15 combinations)\u2014whereas longer horizons showed unstable trajectories. Notably, the highest single-point AUROC model (Masked CNN, 0.961) had the worst deployment profile (NNE 17.6\u2013564), while ITE-Transformer (AUROC 0.924) achieved the most favourable alert burden (NNE 1.5\u20132.4). Deployment-oriented evaluation should complement conventional metrics for continuous monitoring models.\n                  <\/jats:p>","DOI":"10.1038\/s41746-026-02722-2","type":"journal-article","created":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T09:08:45Z","timestamp":1778490525000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Deep learning models for acute kidney injury prediction: multi-center external validation and evaluation under simulated continuous monitoring conditions"],"prefix":"10.1038","volume":"9","author":[{"given":"Kyung Hyun","family":"Lee","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Donghwee","family":"Yoon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hyunsun","family":"Lim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ki-Byung","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yong Kyu","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,8]]},"reference":[{"key":"2722_CR1","doi-asserted-by":"publisher","first-page":"607","DOI":"10.1038\/s41581-018-0052-0","volume":"14","author":"EAJ Hoste","year":"2018","unstructured":"Hoste, E. A. J. et al. Global epidemiology and outcomes of acute kidney injury. Nat. Rev. Nephrol. 14, 607\u2013625 (2018).","journal-title":"Nat. Rev. Nephrol."},{"key":"2722_CR2","doi-asserted-by":"publisher","first-page":"1949","DOI":"10.1016\/S0140-6736(19)32563-2","volume":"394","author":"C Ronco","year":"2019","unstructured":"Ronco, C., Bellomo, R. & Kellum, J. A. Acute kidney injury. Lancet 394, 1949\u20131964 (2019).","journal-title":"Lancet"},{"key":"2722_CR3","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1056\/NEJMra1214243","volume":"371","author":"LS Chawla","year":"2014","unstructured":"Chawla, L. S., Eggers, P. W., Star, R. A. & Kimmel, P. L. Acute kidney injury and chronic kidney disease as interconnected syndromes. N. Engl. J. Med. 371, 58\u201366 (2014).","journal-title":"N. Engl. J. Med."},{"key":"2722_CR4","doi-asserted-by":"publisher","first-page":"193","DOI":"10.1038\/s41581-019-0247-z","volume":"16","author":"MT James","year":"2020","unstructured":"James, M. T., Bhatt, M., Pannu, N. & Tonelli, M. Long-term outcomes of acute kidney injury and strategies for improved care. Nat. Rev. Nephrol. 16, 193\u2013205 (2020).","journal-title":"Nat. Rev. Nephrol."},{"key":"2722_CR5","doi-asserted-by":"publisher","first-page":"e0287398","DOI":"10.1371\/journal.pone.0287398","volume":"18","author":"F Alfieri","year":"2023","unstructured":"Alfieri, F. et al. Continuous and early prediction of future moderate and severe Acute Kidney Injury in critically ill patients: development and multi-centric, multi-national external validation of a machine-learning model. PLoS ONE 18, e0287398 (2023).","journal-title":"PLoS ONE"},{"key":"2722_CR6","doi-asserted-by":"publisher","first-page":"1935","DOI":"10.2215\/CJN.00280116","volume":"11","author":"JL Koyner","year":"2016","unstructured":"Koyner, J. L., Adhikari, R., Edelson, D. P. & Churpek, M. M. Development of a multicenter ward-based AKI prediction model. Clin. J. Am. Soc. Nephrol. 11, 1935\u20131943 (2016).","journal-title":"Clin. J. Am. Soc. Nephrol."},{"key":"2722_CR7","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1038\/s41586-019-1390-1","volume":"572","author":"N Toma\u0161ev","year":"2019","unstructured":"Toma\u0161ev, N. et al. A clinically applicable approach to continuous prediction of future acute kidney injury. Nature 572, 116\u2013119 (2019).","journal-title":"Nature"},{"key":"2722_CR8","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1038\/s41746-020-00346-8","volume":"3","author":"N Rank","year":"2020","unstructured":"Rank, N. et al. Deep-learning-based real-time prediction of acute kidney injury outperforms human predictive performance. npj Digit. Med. 3, 139 (2020).","journal-title":"npj Digit. Med."},{"key":"2722_CR9","doi-asserted-by":"publisher","first-page":"e2012892","DOI":"10.1001\/jamanetworkopen.2020.12892","volume":"3","author":"MM Churpek","year":"2020","unstructured":"Churpek, M. M. et al. Internal and external validation of a machine learning risk score for acute kidney injury. JAMA Netw. Open 3, e2012892 (2020).","journal-title":"JAMA Netw. Open"},{"key":"2722_CR10","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-55629-5","volume":"16","author":"Y Zhang","year":"2025","unstructured":"Zhang, Y. et al. Development and validation of a real-time prediction model for acute kidney injury in hospitalized patients. Nat. Commun. 16, 68 (2025).","journal-title":"Nat. Commun."},{"key":"2722_CR11","first-page":"565","volume":"2017","author":"P Cheng","year":"2017","unstructured":"Cheng, P., Waitman, L. R., Hu, Y. & Liu, M. Predicting inpatient acute kidney injury over different time horizons: How early and accurate? AMIA Annu Symp. Proc. 2017, 565\u2013574 (2017).","journal-title":"AMIA Annu Symp. Proc."},{"key":"2722_CR12","doi-asserted-by":"publisher","first-page":"2266","DOI":"10.1093\/ckj\/sfac181","volume":"15","author":"I Vagliano","year":"2022","unstructured":"Vagliano, I. et al. Machine learning models for predicting acute kidney injury: a systematic review and critical appraisal. Clin. Kidney J. 15, 2266\u20132280 (2022).","journal-title":"Clin. Kidney J."},{"key":"2722_CR13","first-page":"9872","volume":"27","author":"MR Bacci","year":"2023","unstructured":"Bacci, M. R., Minczuk, C. V. B. & Fonseca, F. L. A. A systematic review of artificial intelligence algorithms for predicting acute kidney injury. Eur. Rev. Med. Pharm. Sci. 27, 9872\u20139879 (2023).","journal-title":"Eur. Rev. Med. Pharm. Sci."},{"key":"2722_CR14","doi-asserted-by":"publisher","first-page":"499","DOI":"10.7861\/clinmed.2022-news-concept","volume":"22","author":"B Williams","year":"2022","unstructured":"Williams, B. The National Early Warning Score: from concept to NHS implementation. Clin. Med. 22, 499\u2013505 (2022).","journal-title":"Clin. Med."},{"key":"2722_CR15","first-page":"e0967","volume":"5","author":"A Wong","year":"2023","unstructured":"Wong, A. et al. Facilitators and barriers to interacting with clinical decision support in the ICU: a mixed-methods approach. Crit. Care Explor 5, e0967 (2023).","journal-title":"Crit. Care Explor"},{"key":"2722_CR16","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-017-0430-8","volume":"17","author":"JS Ancker","year":"2017","unstructured":"Ancker, J. S. et al. Effects of workload, work complexity, and repeated alerts on alert fatigue in a clinical decision support system. BMC Med. Inf. Decis. Mak. 17, 36 (2017).","journal-title":"BMC Med. Inf. Decis. Mak."},{"key":"2722_CR17","doi-asserted-by":"publisher","first-page":"756","DOI":"10.1016\/S0140-6736(11)61454-2","volume":"380","author":"R Bellomo","year":"2012","unstructured":"Bellomo, R., Kellum, J. A. & Ronco, C. Acute kidney injury. Lancet 380, 756\u2013766 (2012).","journal-title":"Lancet"},{"key":"2722_CR18","doi-asserted-by":"publisher","first-page":"217","DOI":"10.1038\/nrneph.2017.184","volume":"14","author":"JA Kellum","year":"2018","unstructured":"Kellum, J. A. & Prowle, J. R. Paradigms of acute kidney injury in the intensive care setting. Nat. Rev. Nephrol. 14, 217\u2013230 (2018).","journal-title":"Nat. Rev. Nephrol."},{"key":"2722_CR19","doi-asserted-by":"publisher","first-page":"735","DOI":"10.1038\/nrneph.2012.197","volume":"8","author":"M Haase","year":"2012","unstructured":"Haase, M., Kellum, J. A. & Ronco, C. Subclinical AKI\u2013an emerging syndrome with important consequences. Nat. Rev. Nephrol. 8, 735\u2013739 (2012).","journal-title":"Nat. Rev. Nephrol."},{"key":"2722_CR20","doi-asserted-by":"publisher","first-page":"e66568","DOI":"10.2196\/66568","volume":"27","author":"NJ Cho","year":"2025","unstructured":"Cho, N. J. et al. Machine learning to assist in managing acute kidney injury in general wards: multicenter retrospective study. J. Med. Internet Res. 27, e66568 (2025).","journal-title":"J. Med. Internet Res."},{"key":"2722_CR21","doi-asserted-by":"publisher","first-page":"1070","DOI":"10.1097\/CCM.0000000000003123","volume":"46","author":"JL Koyner","year":"2018","unstructured":"Koyner, J. L., Carey, K. A., Edelson, D. P. & Churpek, M. M. The development of a machine learning inpatient acute kidney injury prediction model. Crit. Care Med. 46, 1070\u20131077 (2018).","journal-title":"Crit. Care Med."},{"key":"2722_CR22","doi-asserted-by":"publisher","first-page":"766","DOI":"10.2215\/CJN.0000000695","volume":"20","author":"JL Koyner","year":"2025","unstructured":"Koyner, J. L. et al. Multicenter development and validation of a multimodal deep learning model to predict moderate to severe AKI. Clin. J. Am. Soc. Nephrol. 20, 766\u2013778 (2025).","journal-title":"Clin. J. Am. Soc. Nephrol."},{"key":"2722_CR23","doi-asserted-by":"publisher","first-page":"ooaf065","DOI":"10.1093\/jamiaopen\/ooaf065","volume":"8","author":"T Shi","year":"2025","unstructured":"Shi, T., Lin, Y., Zhao, H. & Kong, G. Artificial intelligence models for predicting acute kidney injury in the intensive care unit: a systematic review of modeling methods, data utilization, and clinical applicability. JAMIA Open 8, ooaf065 (2025).","journal-title":"JAMIA Open"},{"key":"2722_CR24","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","volume":"385","author":"GS Collins","year":"2024","unstructured":"Collins, G. S. et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ 385, e078378 (2024).","journal-title":"BMJ"},{"key":"2722_CR25","doi-asserted-by":"publisher","first-page":"146045822110075","DOI":"10.1177\/14604582211007536","volume":"27","author":"OO Olakotan","year":"2021","unstructured":"Olakotan, O. O. & Mohd Yusof, M. The appropriateness of clinical decision support systems alerts in supporting clinical workflows: a systematic review. Health Inform. J. 27, 14604582211007536 (2021).","journal-title":"Health Inform. J."},{"key":"2722_CR26","doi-asserted-by":"publisher","DOI":"10.1186\/s12916-019-1466-7","volume":"17","author":"B Van Calster","year":"2019","unstructured":"Van Calster, B., McLernon, D. J., van Smeden, M., Wynants, L. & Steyerberg, E. W. Calibration: the Achilles heel of predictive analytics. BMC Med. 17, 230 (2019).","journal-title":"BMC Med."},{"key":"2722_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-022-01899-x","volume":"10","author":"AEW Johnson","year":"2023","unstructured":"Johnson, A. E. W. et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci. Data 10, 1 (2023).","journal-title":"Sci. Data"},{"key":"2722_CR28","doi-asserted-by":"publisher","first-page":"c179","DOI":"10.1159\/000339789","volume":"120","author":"A Khwaja","year":"2012","unstructured":"Khwaja, A. KDIGO clinical practice guidelines for acute kidney injury. Nephron Clin. Pract. 120, c179\u2013c184 (2012).","journal-title":"Nephron Clin. Pract."},{"key":"2722_CR29","first-page":"607","volume":"327","author":"KH Lee","year":"2025","unstructured":"Lee, K. H., Hahn, S., Lim, H., Lee, K. B. & Lee, Y. K. Development of labeling algorithm for early prediction of acute kidney injury. Stud. Health Technol. Inf. 327, 607\u2013608 (2025).","journal-title":"Stud. Health Technol. Inf."},{"key":"2722_CR30","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-024-02514-2","volume":"24","author":"VK Chauhan","year":"2024","unstructured":"Chauhan, V. K. et al. Continuous patient state attention model for addressing irregularity in electronic health records. BMC Med. Inf. Decis. Mak. 24, 117 (2024).","journal-title":"BMC Med. Inf. Decis. Mak."},{"key":"2722_CR31","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-24271-9","volume":"8","author":"Z Che","year":"2018","unstructured":"Che, Z., Purushotham, S., Cho, K., Sontag, D. & Liu, Y. Recurrent neural networks for multivariate time series with missing values. Sci. Rep. 8, 6085 (2018).","journal-title":"Sci. Rep."},{"key":"2722_CR32","doi-asserted-by":"publisher","first-page":"1672260","DOI":"10.3389\/fpsyt.2025.1672260","volume":"16","author":"KH Lee","year":"2025","unstructured":"Lee, K. H., Cho, C. H., Kim, A. Y., Jeon, H. J. & Byun, S. Deep learning-based stress detection from RR intervals in major depressive disorder, panic disorder, and healthy individuals. Front Psychiatry 16, 1672260 (2025).","journal-title":"Front Psychiatry"},{"key":"2722_CR33","doi-asserted-by":"publisher","first-page":"Article 105","DOI":"10.1145\/3516367","volume":"16","author":"S Tipirneni","year":"2022","unstructured":"Tipirneni, S. & Reddy, C. K. Self-supervised transformer for sparse and irregularly sampled multivariate clinical time-series. ACM Trans. Knowl. Discov. Data 16, Article 105 (2022).","journal-title":"ACM Trans. Knowl. Discov. Data"},{"key":"2722_CR34","doi-asserted-by":"publisher","first-page":"1322","DOI":"10.1093\/jamia\/ocae088","volume":"31","author":"FS Bashiri","year":"2024","unstructured":"Bashiri, F. S. et al. Development and external validation of deep learning clinical prediction models using variable-length time series data. J. Am. Med. Inf. Assoc. 31, 1322\u20131330 (2024).","journal-title":"J. Am. Med. Inf. Assoc."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02722-2","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02722-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02722-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:03:23Z","timestamp":1784073803000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02722-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,8]]},"references-count":34,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2722"],"URL":"https:\/\/doi.org\/10.1038\/s41746-026-02722-2","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,8]]},"assertion":[{"value":"23 February 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"28 April 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 May 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":"Ethics"}}],"article-number":"544"}}