{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,31]],"date-time":"2026-03-31T04:39:07Z","timestamp":1774931947867,"version":"3.50.1"},"reference-count":23,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T00:00:00Z","timestamp":1752537600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["OT2OD032701"],"award-info":[{"award-number":["OT2OD032701"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01GM139967"],"award-info":[{"award-number":["R01GM139967"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Early identification of sepsis in the emergency department using machine learning remains a challenging problem, primarily due to the lack of a gold standard for sepsis diagnosis, the heterogeneity in clinical presentations, and the impact of confounding conditions.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>In this work, we present a deep-learning-based predictive model designed to enable early detection of patients at risk of developing sepsis, using data from the first 24 h of admission. The model is based on routine blood test results commonly performed on patients, including CBC (Complete Blood Count), CMP (Comprehensive Metabolic Panel), lipid panels, vital signs, age, and sex. To address the challenge of label uncertainty as a part of the training process, we explore two different definitions, namely, Sepsis-3 and Adult Sepsis Event. We analyze the advantages and limitations of each in the context of patient clinical parameters and comorbidities. We specifically examine how the quality of the ground truth label influences the performance of the deep learning system and evaluate the effect of a consensus-based approach that incorporates both definitions. We also evaluated the model's performance across sub-cohorts, including patients with confounding comorbidities (such as chronic kidney, liver disease, and coagulation disorders) and those with infections confirmed by billing codes.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Our results show that the consensus-based model identifies at-risk patients in the first 24 h with 83.7% sensitivity, 80% specificity, 36% PPV, 97% NPV, and an AUC of 0.9. Our cohort-wise analysis revealed a high PPV (77%) in infection-confirmed subgroups and a drop in specificity across cohorts with confounding comorbidities (47-70%).<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>This work highlights the limitations of retrospective sepsis definitions and underscores the need for tailored approaches in automated sepsis detection, particularly when dealing with patients with confounding comorbidities.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frai.2025.1452471","type":"journal-article","created":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T12:43:47Z","timestamp":1752583427000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Evaluating the impact of common clinical confounders on performance of deep-learning-based sepsis risk assessment"],"prefix":"10.3389","volume":"8","author":[{"given":"Shikha","family":"Chaganti","sequence":"first","affiliation":[]},{"given":"Vivek","family":"Singh","sequence":"additional","affiliation":[]},{"given":"Alasdair Edward","family":"Gent","sequence":"additional","affiliation":[]},{"given":"Rishikesan","family":"Kamaleswaran","sequence":"additional","affiliation":[]},{"given":"Ali","family":"Kamen","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,7,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"e31083","DOI":"10.2196\/31083","article-title":"Computerized clinical decision support systems for the early detection of sepsis among adult inpatients: scoping review","volume":"24","author":"Ackermann","year":"2022","journal-title":"J. Med. Internet Res"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1644","DOI":"10.1378\/chest.101.6.1644","article-title":"Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis","volume":"101","author":"Bone","year":"1992","journal-title":"Chest"},{"key":"B3","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1007\/s00134-019-05872-y","article-title":"Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy","volume":"46","author":"Fleuren","year":"2020","journal-title":"Intensive Care Med"},{"key":"B4","doi-asserted-by":"publisher","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":"B5","doi-asserted-by":"publisher","first-page":"544","DOI":"10.1016\/j.annemergmed.2017.01.008","article-title":"An emergency department validation of the sep-3 sepsis and septic shock definitions and comparison with 1992 consensus definitions","volume":"70","author":"Henning","year":"2017","journal-title":"Ann. Emerg. Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"e0053","DOI":"10.1097\/CCE.0000000000000053","article-title":"Comparison of automated sepsis identification methods and electronic health record-based sepsis phenotyping: improving case identification accuracy by accounting for confounding comorbid conditions","volume":"1","author":"Henry","year":"2019","journal-title":"Crit. Care Explor"},{"key":"B7","unstructured":"Johnson\n              A.\n            \n            \n              Bulgarelli\n              L.\n            \n            \n              Pollard\n              T.\n            \n            \n              Horng\n              S.\n            \n            \n              Celi\n              L. A.\n            \n            \n              Mark\n              R.\n            \n          \n          Mimic-iv. PhysioNet\n          \n          2020"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-023-01945-2","article-title":"Mimic-iv, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci. Data"},{"key":"B9","doi-asserted-by":"publisher","first-page":"483","DOI":"10.1007\/s12021-021-09553-4","article-title":"pyphewas: a phenome-disease association tool for electronic medical record analysis","volume":"20","author":"Kerley","year":"2022","journal-title":"Neuroinformatics"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1250","DOI":"10.1097\/01.CCM.0000050454.01978.3B","article-title":"2001 sccm\/esicm\/accp\/ats\/sis international sepsis definitions conference","volume":"31","author":"Levy","year":"2003","journal-title":"Crit. Care Med"},{"key":"B11","doi-asserted-by":"publisher","first-page":"745","DOI":"10.1111\/acem.14265","article-title":"Most emergency department patients meeting sepsis criteria are not diagnosed with sepsis at discharge","volume":"28","author":"Litell","year":"2021","journal-title":"Acad. Emerg. Med"},{"key":"B12","doi-asserted-by":"publisher","first-page":"887","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":"B13","article-title":"\u201cA unified approach to interpreting model predictions,\u201d","author":"Lundberg","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B14","article-title":"Umap: Uniform manifold approximation and projection for dimension reduction","author":"McInnes","year":"2018","journal-title":"arXiv preprint arXiv:1802.03426"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1145\/1102351.1102430","article-title":"\u201cPredicting good probabilities with supervised learning,\u201d","author":"Niculescu-Mizil","year":"2005","journal-title":"Proceedings of the 22nd International Conference on Machine Learning"},{"key":"B16","doi-asserted-by":"publisher","first-page":"2234","DOI":"10.1097\/01.CCM.0000145586.23276.0F","article-title":"Epidemiology of sepsis in patients with traumatic injury","volume":"32","author":"Osborn","year":"2004","journal-title":"Crit. Care Med"},{"key":"B17","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1136\/bmjqs-2018-008331","article-title":"Using objective clinical data to track progress on preventing and treating sepsis: CDC's new \u2018Adult sepsis event' surveillance strategy","volume":"28","author":"Rhee","year":"2019","journal-title":"BMJ Qual. Saf."},{"key":"B18","doi-asserted-by":"publisher","first-page":"272","DOI":"10.1136\/emermed-2020-211111","article-title":"Retrospective single-centre descriptive study of the characteristics, management and outcomes of adult patients with suspected sepsis in the emergency department","volume":"39","author":"Sabir","year":"2022","journal-title":"Emerg. Med. J"},{"key":"B19","doi-asserted-by":"publisher","first-page":"801","DOI":"10.1001\/jama.2016.0287","article-title":"The third international consensus definitions for sepsis and septic shock (sepsis-3)","volume":"315","author":"Singer","year":"2016","journal-title":"JAMA"},{"key":"B20","doi-asserted-by":"publisher","first-page":"103523","DOI":"10.1016\/j.isci.2021.103523","article-title":"A deep learning approach for predicting severity of covid-19 patients using a parsimonious set of laboratory markers","volume":"24","author":"Singh","year":"2021","journal-title":"Iscience"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1349","DOI":"10.1093\/jamia\/ocad075","article-title":"Deployment of machine learning algorithms to predict sepsis: systematic review and application of the salient clinical ai implementation framework","volume":"30","author":"van der Vegt","year":"2023","journal-title":"J. Am. Med. Inform. Assoc"},{"key":"B22","doi-asserted-by":"publisher","first-page":"e14325","DOI":"10.2196\/14325","article-title":"Mapping icd-10 and icd-10-cm codes to phecodes: workflow development and initial evaluation","volume":"7","author":"Wu","year":"2019","journal-title":"JMIR Med. Inform"},{"key":"B23","doi-asserted-by":"publisher","first-page":"689","DOI":"10.1016\/S0272-6386(01)80116-5","article-title":"Hypocalcemia: a pervasive metabolic abnormality in the critically ill","volume":"37","author":"Zivin","year":"2001","journal-title":"Am. J. Kidney Dis"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1452471\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T12:43:48Z","timestamp":1752583428000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1452471\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,15]]},"references-count":23,"alternative-id":["10.3389\/frai.2025.1452471"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1452471","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,15]]},"article-number":"1452471"}}