{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T04:17:04Z","timestamp":1776313024308,"version":"3.50.1"},"reference-count":21,"publisher":"Oxford University Press (OUP)","issue":"Supplement_1","license":[{"start":{"date-parts":[[2024,6,28]],"date-time":"2024-06-28T00:00:00Z","timestamp":1719532800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001711","name":"Swiss National Science Foundation","doi-asserted-by":"publisher","award":["205321_176005"],"award-info":[{"award-number":["205321_176005"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Strategic Focus Area \u201cPersonalized Health and Related Technologies"},{"name":"Swiss Federal Institutes of Technology","award":["#2022\u2013278"],"award-info":[{"award-number":["#2022\u2013278"]}]},{"name":"European Union\u2019s Horizon 2020"},{"name":"Marie Sklodowska-Curie","award":["813533"],"award-info":[{"award-number":["813533"]}]},{"DOI":"10.13039\/501100004189","name":"Max Planck Society","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004189","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,6,28]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Acute kidney injury (AKI) is a syndrome that affects a large fraction of all critically ill patients, and early diagnosis to receive adequate treatment is as imperative as it is challenging to make early. Consequently, machine learning approaches have been developed to predict AKI ahead of time. However, the prevalence of AKI is often underestimated in state-of-the-art approaches, as they rely on an AKI event annotation solely based on creatinine, ignoring urine output.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\u2002<\/jats:title>\n                    <jats:p>We construct and evaluate early warning systems for AKI in a multi-disciplinary ICU setting, using the complete KDIGO definition of AKI. We propose several variants of gradient-boosted decision tree (GBDT)-based models, including a novel time-stacking based approach. A state-of-the-art LSTM-based model previously proposed for AKI prediction is used as a comparison, which was not specifically evaluated in ICU settings yet.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We find that optimal performance is achieved by using GBDT with the time-based stacking technique (AUPRC\u2009=\u200965.7%, compared with the LSTM-based model\u2019s AUPRC\u2009=\u200962.6%), which is motivated by the high relevance of time since ICU admission for this task. Both models show mildly reduced performance in the limited training data setting, perform fairly across different subcohorts, and exhibit no issues in gender transfer.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\u2002<\/jats:title>\n                    <jats:p>Following the official KDIGO definition substantially increases the number of annotated AKI events. In our study GBDTs outperform LSTM models for AKI prediction. Generally, we find that both model types are robust in a variety of challenging settings arising for ICU data.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>The code to reproduce the findings of our manuscript can be found at: https:\/\/github.com\/ratschlab\/AKI-EWS<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btae212","type":"journal-article","created":{"date-parts":[[2024,4,15]],"date-time":"2024-04-15T11:52:37Z","timestamp":1713181957000},"page":"i247-i256","source":"Crossref","is-referenced-by-count":11,"title":["An empirical study on KDIGO-defined acute kidney injury prediction in the intensive care unit"],"prefix":"10.1093","volume":"40","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-2519-0370","authenticated-orcid":false,"given":"Xinrui","family":"Lyu","sequence":"first","affiliation":[{"name":"Department of Computer Science, ETH Z\u00fcrich , Z\u00fcrich, 8092, Switzerland"},{"name":"NEXUS Personalized Health Technologies, ETH Z\u00fcrich , Schlieren, 8952, Switzerland"},{"name":"Swiss Institute for Bioinformatics , Lausanne, 1015, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5005-9209","authenticated-orcid":false,"given":"Bowen","family":"Fan","sequence":"additional","affiliation":[{"name":"Swiss Institute for Bioinformatics , Lausanne, 1015, Switzerland"},{"name":"Department of Biosystems Science and Engineering, ETH Z\u00fcrich , Basel, 4056, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6397-1689","authenticated-orcid":false,"given":"Matthias","family":"H\u00fcser","sequence":"additional","affiliation":[{"name":"Department of Computer Science, ETH Z\u00fcrich , Z\u00fcrich, 8092, Switzerland"},{"name":"Swiss Institute for Bioinformatics , Lausanne, 1015, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1067-7651","authenticated-orcid":false,"given":"Philip","family":"Hartout","sequence":"additional","affiliation":[{"name":"Department of Biosystems Science and Engineering, ETH Z\u00fcrich , Basel, 4056, Switzerland"},{"name":"Department of Machine Learning and Systems Biology, Max Planck Institute of Biochemistry , Martinsried, 82152, Germany"}]},{"given":"Thomas","family":"Gumbsch","sequence":"additional","affiliation":[{"name":"Swiss Institute for Bioinformatics , Lausanne, 1015, Switzerland"},{"name":"Department of Biosystems Science and Engineering, ETH Z\u00fcrich , Basel, 4056, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5579-810X","authenticated-orcid":false,"given":"Martin","family":"Faltys","sequence":"additional","affiliation":[{"name":"Department of Intensive Care, Austin Hospital , Melbourne, Victoria, 3084, Australia"},{"name":"Department of Intensive Care Medicine, University Hospital, University of Bern , Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3149-5873","authenticated-orcid":false,"given":"Tobias M","family":"Merz","sequence":"additional","affiliation":[{"name":"Cardiovascular Intensive Care Unit, Auckland City Hospital , Auckland, 1023, New Zealand"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5486-8532","authenticated-orcid":false,"given":"Gunnar","family":"R\u00e4tsch","sequence":"additional","affiliation":[{"name":"Department of Computer Science, ETH Z\u00fcrich , Z\u00fcrich, 8092, Switzerland"},{"name":"Swiss Institute for Bioinformatics , Lausanne, 1015, Switzerland"},{"name":"Medical Informatics Unit, Z\u00fcrich University Hospital , 8091, Switzerland"},{"name":"AI Center at ETH Z\u00fcrich , Z\u00fcrich, 8092, Switzerland"},{"name":"Department of Biology, ETH Z\u00fcrich , Z\u00fcrich, 8093, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7221-2393","authenticated-orcid":false,"given":"Karsten","family":"Borgwardt","sequence":"additional","affiliation":[{"name":"Swiss Institute for Bioinformatics , Lausanne, 1015, Switzerland"},{"name":"Department of Biosystems Science and Engineering, ETH Z\u00fcrich , Basel, 4056, Switzerland"},{"name":"Department of Machine Learning and Systems Biology, Max Planck Institute of Biochemistry , Martinsried, 82152, Germany"}]}],"member":"286","published-online":{"date-parts":[[2024,6,28]]},"reference":[{"key":"2024070715412017500_btae212-B1","doi-asserted-by":"crossref","first-page":"147","DOI":"10.2215\/CJN.12191213","article-title":"Biomarkers of AKI: a review of mechanistic relevance and potential therapeutic implications","volume":"10","author":"Alge","year":"2015","journal-title":"Clin J Am Soc Nephrol"},{"key":"2024070715412017500_btae212-B2","doi-asserted-by":"crossref","first-page":"1121","DOI":"10.1038\/s42256-022-00563-8","article-title":"Generalizability of an acute kidney injury prediction model across health systems","volume":"4","author":"Cao","year":"2022","journal-title":"Nat Mach Intell"},{"key":"2024070715412017500_btae212-B3","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1038\/s43856-022-00225-1","article-title":"Identifying and evaluating barriers for the implementation of machine learning in the intensive care unit","volume":"2","author":"D'Hondt","year":"2022","journal-title":"Commun Med (Lond)"},{"key":"2024070715412017500_btae212-B4","doi-asserted-by":"crossref","first-page":"e215","DOI":"10.1161\/01.CIR.101.23.e215","article-title":"Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals","volume":"101","author":"Goldberger","year":"2000","journal-title":"Circulation"},{"key":"2024070715412017500_btae212-B5","first-page":"507","volume-title":"Adv Neural Inf process Syst","author":"Grinsztajn","year":"2022"},{"key":"2024070715412017500_btae212-B6","doi-asserted-by":"crossref","first-page":"19738","DOI":"10.1038\/s41598-021-99141-y","article-title":"Perioperative doppler measurements of renal perfusion are associated with acute kidney injury in patients undergoing cardiac surgery","volume":"11","author":"Hermansen","year":"2021","journal-title":"Sci Rep"},{"key":"2024070715412017500_btae212-B7","author":"H\u00fcser","year":"2024","journal-title":"A Comprehensive ML-Based Respiratory Monitoring System for Physiological Monitoring & Resource Planning in the ICU"},{"key":"2024070715412017500_btae212-B8","doi-asserted-by":"crossref","first-page":"364","DOI":"10.1038\/s41591-020-0789-4","article-title":"Early prediction of circulatory failure in the intensive care unit using machine learning","volume":"26","author":"Hyland","year":"2020","journal-title":"Nat Med"},{"key":"2024070715412017500_btae212-B9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci Data"},{"key":"2024070715412017500_btae212-B10","article-title":"LightGBM: a highly efficient gradient boosting decision tree","volume":"30","author":"Ke","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"2024070715412017500_btae212-B11","first-page":"19","article-title":"KDIGO clinical practice guideline for acute kidney injury. Section 2: AKI Definition","volume":"2","author":"Kellum","year":"2012","journal-title":"Kidney Int Suppl"},{"key":"2024070715412017500_btae212-B12","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1038\/s41572-021-00284-z","article-title":"Acute kidney injury","volume":"7","author":"Kellum","year":"2021","journal-title":"Nat Rev Dis Primers"},{"key":"2024070715412017500_btae212-B13","first-page":"265","article-title":"Association between e-alert implementation for detection of acute kidney injury and outcomes: a systematic review","volume":"32","author":"Lachance","year":"2017","journal-title":"Nephrol Dial Transplant"},{"key":"2024070715412017500_btae212-B14","article-title":"A unified approach to interpreting model predictions","volume":"30","author":"Lundberg","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"2024070715412017500_btae212-B15","doi-asserted-by":"crossref","first-page":"843","DOI":"10.1093\/bja\/aes357","article-title":"Novel biomarkers of acute kidney injury and failure: clinical applicability","volume":"109","author":"M\u00e5rtensson","year":"2012","journal-title":"Br J Anaesth"},{"key":"2024070715412017500_btae212-B16","doi-asserted-by":"crossref","first-page":"2054358118776326","DOI":"10.1177\/2054358118776326","article-title":"Prediction of acute kidney injury with a machine learning algorithm using electronic health record data","volume":"5","author":"Mohamadlou","year":"2018","journal-title":"Can J Kidney Health Dis"},{"key":"2024070715412017500_btae212-B17","doi-asserted-by":"crossref","first-page":"142","DOI":"10.1038\/s41746-022-00679-6","article-title":"Discrete-time survival analysis in the critically ill: a deep learning approach using heterogeneous data","volume":"5","author":"Thorsen-Meyer","year":"2022","journal-title":"NPJ Digit Med"},{"key":"2024070715412017500_btae212-B18","doi-asserted-by":"crossref","first-page":"116","DOI":"10.1038\/s41586-019-1390-1","article-title":"A clinically applicable approach to continuous prediction of future acute kidney injury","volume":"572","author":"Toma\u0161ev","year":"2019","journal-title":"Nature"},{"key":"2024070715412017500_btae212-B19","doi-asserted-by":"crossref","first-page":"399","DOI":"10.1016\/j.ress.2015.05.018","article-title":"Variable importance analysis: a comprehensive review","volume":"142","author":"Wei","year":"2015","journal-title":"Reliability Eng Syst Saf"},{"key":"2024070715412017500_btae212-B20","doi-asserted-by":"crossref","first-page":"160018","DOI":"10.1038\/sdata.2016.18","article-title":"The fair guiding principles for scientific data management and stewardship","volume":"3","author":"Wilkinson","year":"2016","journal-title":"Sci Data"},{"key":"2024070715412017500_btae212-B21","author":"Yeche","year":"2021"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/40\/Supplement_1\/i247\/58464357\/btae212.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/40\/Supplement_1\/i247\/58464357\/btae212.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,7,7]],"date-time":"2024-07-07T11:41:54Z","timestamp":1720352514000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/40\/Supplement_1\/i247\/7700890"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,6,28]]},"references-count":21,"journal-issue":{"issue":"Supplement_1","published-print":{"date-parts":[[2024,6,28]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btae212","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2024.02.01.24302063","asserted-by":"object"}]},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2024,7]]},"published":{"date-parts":[[2024,6,28]]}}}