{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,23]],"date-time":"2026-08-23T13:56:35Z","timestamp":1787493395801,"version":"build-2736575974"},"reference-count":42,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,6,28]],"date-time":"2023-06-28T00:00:00Z","timestamp":1687910400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"NIH","doi-asserted-by":"publisher","award":["\u00a0"],"award-info":[{"award-number":["\u00a0"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000071","name":"NICHD","doi-asserted-by":"publisher","award":["5T32HD040686"],"award-info":[{"award-number":["5T32HD040686"]}],"id":[{"id":"10.13039\/100000071","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"NIH","doi-asserted-by":"publisher","award":["\u00a0"],"award-info":[{"award-number":["\u00a0"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000071","name":"NICHD","doi-asserted-by":"publisher","award":["1K23HD099331-01A1"],"award-info":[{"award-number":["1K23HD099331-01A1"]}],"id":[{"id":"10.13039\/100000071","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Pediatr."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Acute neurological injury is a leading cause of permanent disability and death in the pediatric intensive care unit (PICU). No predictive model has been validated for critically ill children with acute neurological injury.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>We hypothesized that PICU patients with concern for acute neurological injury are at higher risk for morbidity and mortality, and advanced analytics would derive robust, explainable subgroup models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We performed a secondary subgroup analysis of the Trichotomous Outcomes in Pediatric Critical Care (TOPICC) study (2011\u20132013), predicting mortality and morbidity from admission physiology (lab values and vital signs in 6\u2005h surrounding admission). We analyzed patients with suspected acute neurological injury using standard machine learning algorithms. Feature importance was analyzed using SHapley Additive exPlanations (SHAP). We created a Fast Healthcare Interoperability Resources (FHIR) application to demonstrate potential for interoperability using pragmatic data.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>1,860 patients had suspected acute neurological injury at PICU admission, with higher morbidity (8.2 vs. 3.4%) and mortality (6.2 vs. 1.9%) than those without similar concern. The ensemble regressor (containing Random Forest, Gradient Boosting, and Support Vector Machine learners) produced the best model, with Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.91 [95% CI (0.88, 0.94)] and Average Precision (AP) of 0.59 [0.51, 0.69] for mortality, and decreased performance predicting simultaneous mortality and morbidity (0.83 [0.80, 0.86] and 0.59 [0.51, 0.64]); at a set specificity of 0.995, positive predictive value (PPV) was 0.79 for mortality, and 0.88 for mortality and morbidity. By comparison, for mortality, the TOPICC logistic regression had AUROC of 0.90 [0.84, 0.93], but substantially inferior AP of 0.49 [0.35, 0.56] and PPV of 0.60 at specificity 0.995. Feature importance analysis showed that pupillary non-reactivity, Glasgow Coma Scale, and temperature were the most contributory vital signs, and acidosis and coagulopathy the most important laboratory values. The FHIR application provided a simulated demonstration of real-time health record query and model deployment.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>PICU patients with suspected acute neurological injury have higher mortality and morbidity. Our machine learning approach independently identified previously-known causes of secondary brain injury. Advanced modeling achieves improved positive predictive value in this important population compared to published models, providing a stepping stone in the path to deploying explainable models as interoperable bedside decision-support tools.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fped.2023.1177470","type":"journal-article","created":{"date-parts":[[2023,6,28]],"date-time":"2023-06-28T05:35:30Z","timestamp":1687930530000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":12,"title":["Interoperable and explainable machine learning models to predict morbidity and mortality in acute neurological injury in the pediatric intensive care unit: secondary analysis of the TOPICC study"],"prefix":"10.3389","volume":"11","author":[{"given":"Neil K.","family":"Munjal","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Robert S. B.","family":"Clark","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dennis W.","family":"Simon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Patrick M.","family":"Kochanek","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher M.","family":"Horvat","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2023,6,28]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"566","DOI":"10.1097\/PCC.0b013e3181fe3420","article-title":"Brain injuries and neurological system failure are the most common proximate cause of death in children admitted to a pediatric intensive care unit","volume":"12","author":"Au","year":"2011","journal-title":"Pediatr Crit Care Med"},{"key":"B2","doi-asserted-by":"publisher","first-page":"330","DOI":"10.1097\/PCC.0000000000001093","article-title":"International survey of critically ill children with acute neurologic insults: the prevalence of acute critical neurological disease in children: a global epidemiological assessment study","volume":"18","author":"Fink","year":"2017","journal-title":"Pediatr Crit Care Med"},{"key":"B3","doi-asserted-by":"publisher","first-page":"1737","DOI":"10.1097\/00003246-199810000-00032","article-title":"Differences in pediatric ICU mortality risk over time","volume":"26","author":"Tilford","year":"1998","journal-title":"Crit Care Med"},{"key":"B4","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1186\/1477-7525-6-21","article-title":"Quality of life in children three and nine months after discharge from a paediatric intensive care unit: a prospective cohort study","volume":"6","author":"Knoester","year":"2008","journal-title":"Health Qual Life Outcomes"},{"key":"B5","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1097\/00003246-198202000-00004","article-title":"Predicting mortality of intensive care unit patients. The importance of coma","volume":"10","author":"Teres","year":"1982","journal-title":"Crit Care Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"254","DOI":"10.1097\/MCC.0b013e328344f2ae","article-title":"Predicting neurological outcome after cardiac arrest","volume":"17","author":"Oddo","year":"2011","journal-title":"Curr Opin Crit Care"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/1471-2253-15-1","article-title":"Predictors of neurological outcomes after successful extracorporeal cardiopulmonary resuscitation","volume":"15","author":"Ryu","year":"2015","journal-title":"BMC Anesthesiol"},{"key":"B8","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1097\/CCM.0000000000002753","article-title":"Development and validation of an empiric tool to predict favorable neurologic outcomes among PICU patients","volume":"46","author":"Gupta","year":"2018","journal-title":"Crit Care Med"},{"key":"B9","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1186\/s13054-019-2561-z","article-title":"A deep learning model for real-time mortality prediction in critically ill children","volume":"23","author":"Kim","year":"2019","journal-title":"Crit Care"},{"key":"B10","doi-asserted-by":"publisher","first-page":"641","DOI":"10.1038\/s41390-019-0510-9","article-title":"Modelling outcomes after paediatric brain injury with admission laboratory values: a machine-learning approach","volume":"86","author":"Kayhanian","year":"2019","journal-title":"Pediatr Res"},{"key":"B11","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1097\/CCM.0000000000002936","article-title":"An interpretable machine learning model for accurate prediction of sepsis in the ICU","volume":"46","author":"Nemati","year":"2018","journal-title":"Crit Care Med"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1629","DOI":"10.2307\/1939921","article-title":"Nontraditional regression analyses","volume":"74","author":"Trexler","year":"1993","journal-title":"Ecology"},{"key":"B13","doi-asserted-by":"publisher","first-page":"269","DOI":"10.1097\/PCC.0000000000001737","article-title":"Management of pediatric severe traumatic brain injury: 2019 consensus and guidelines-based algorithm for first and second tier therapies","volume":"20","author":"Kochanek","year":"2019","journal-title":"Pediatr Crit Care Med"},{"key":"B14","doi-asserted-by":"publisher","first-page":"327","DOI":"10.1159\/000094158","article-title":"Serum biomarkers after traumatic and hypoxemic brain injuries: insight into the biochemical response of the pediatric brain to inflicted brain injury","volume":"28","author":"Pardes Berger","year":"2006","journal-title":"Dev Neurosci"},{"key":"B15","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1007\/s12028-017-0414-7","article-title":"Brain-Specific serum biomarkers predict neurological morbidity in diagnostically diverse pediatric intensive care unit patients","volume":"28","author":"Au","year":"2018","journal-title":"Neurocrit Care"},{"key":"B16","doi-asserted-by":"publisher","first-page":"352","DOI":"10.1016\/S1532-0464(03)00034-0","article-title":"Logistic regression and artificial neural network classification models: a methodology review","volume":"35","author":"Dreiseitl","year":"2002","journal-title":"J Biomed Inform"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1699","DOI":"10.1097\/CCM.0000000000001081","article-title":"Simultaneous prediction of new morbidity, mortality, and survival without new morbidity from pediatric intensive care: a new paradigm for outcomes assessment","volume":"43","author":"Pollack","year":"2015","journal-title":"Crit Care Med"},{"key":"B18","doi-asserted-by":"publisher","first-page":"623","DOI":"10.1097\/CCM.0000000000004246","article-title":"Development and reporting of prediction models: guidance for authors from editors of respiratory, sleep, and critical care journals","volume":"48","author":"Leisman","year":"2020","journal-title":"Crit Care Med"},{"key":"B19","doi-asserted-by":"publisher","first-page":"743","DOI":"10.1097\/00003246-199605000-00004","article-title":"PRISM III: an updated pediatric risk of mortality score","volume":"24","author":"Pollack","year":"1996","journal-title":"Crit Care Med"},{"key":"B20","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1097\/PCC.0b013e31829760cf","article-title":"Paediatric index of mortality 3: an updated model for predicting mortality in pediatric intensive care*","volume":"14","author":"Straney","year":"2013","journal-title":"Pediatr Crit Care Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"554633","DOI":"10.3389\/fneur.2020.554633","article-title":"Machine learning applications in the neuro ICU: a solution to big data mayhem?","volume":"11","author":"Chaudhry","year":"2020","journal-title":"Front Neurol"},{"key":"B22","doi-asserted-by":"publisher","first-page":"e18","DOI":"10.1542\/peds.2008-1987","article-title":"Functional status scale: new pediatric outcome measure","volume":"124","author":"Pollack","year":"2009","journal-title":"Pediatrics"},{"key":"B23","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"B24","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1186\/s12916-019-1466-7","article-title":"Calibration: the Achilles heel of predictive analytics","volume":"17","author":"Van Calster","year":"2019","journal-title":"BMC Med"},{"key":"B25","author":"Lundberg","year":""},{"key":"B26","doi-asserted-by":"publisher","first-page":"55","DOI":"10.7326\/M14-0697","article-title":"Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement","volume":"162","author":"Collins","year":"20156","journal-title":"Ann Intern Med"},{"key":"B27","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.jclinepi.2019.02.004","article-title":"A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models","volume":"110","author":"Christodoulou","year":"2019","journal-title":"J Clin Epidemiol"},{"key":"B28","doi-asserted-by":"publisher","first-page":"097603","DOI":"10.1063\/1.4915529","article-title":"Dynamical disease: challenges for nonlinear dynamics and medicine","volume":"25","author":"Glass","year":"2015","journal-title":"Chaos"},{"key":"B29","doi-asserted-by":"publisher","first-page":"270","DOI":"10.1186\/s12859-018-2264-5","article-title":"Random forest versus logistic regression: a large-scale benchmark experiment","volume":"19","author":"Couronn\u00e9","year":"2018","journal-title":"BMC Bioinformatics"},{"key":"B30","first-page":"648","author":"Bhatt","year":"2020"},{"key":"B31","first-page":"55","author":"Holzinger","year":"2018"},{"key":"B32","doi-asserted-by":"publisher","first-page":"3711","DOI":"10.1093\/bioinformatics\/bty373","article-title":"The revival of the gini importance?","volume":"34","author":"Nembrini","year":"2018","journal-title":"Bioinformatics"},{"key":"B33","doi-asserted-by":"publisher","first-page":"101647","DOI":"10.1016\/j.compenvurbsys.2021.101647","article-title":"Peeking inside the black-box: explainable machine learning applied to household transportation energy consumption","volume":"88","author":"Shams Amiri","year":"2021","journal-title":"Comput Environ Urban Syst"},{"key":"B34","doi-asserted-by":"publisher","first-page":"104041","DOI":"10.1016\/j.compbiomed.2020.104041","article-title":"An explainable machine learning model for early detection of Parkinson\u2019s disease using LIME on DaTSCAN imagery","volume":"126","author":"Magesh","year":"2020","journal-title":"Comput Biol Med"},{"key":"B35","doi-asserted-by":"publisher","first-page":"749","DOI":"10.1038\/s41551-018-0304-0","article-title":"Explainable machine-learning predictions for the prevention of hypoxaemia during surgery","volume":"2","author":"Lundberg","year":"2018","journal-title":"Nat Biomed Eng"},{"key":"B36","first-page":"4093","author":"Henry","year":"2018"},{"key":"B37","first-page":"326","author":"Bender","year":"2013"},{"key":"B38","doi-asserted-by":"publisher","first-page":"103310","DOI":"10.1016\/j.jbi.2019.103310","article-title":"Developing a FHIR-based EHR phenotyping framework: a case study for identification of patients with obesity and multiple comorbidities from discharge summaries","volume":"99","author":"Hong","year":"2019","journal-title":"J Biomed Inform"},{"key":"B39","first-page":"25510","article-title":"Medicare and medicaid programs; patient protection and affordable care act; interoperability and patient access for medicare advantage organization and medicaid managed care plans, state medicaid agencies, CHIP agencies and CHIP managed care entities, issuers of qualified health plans in the federally-facilitated exchanges and health care providers","volume":"85","year":"2020","journal-title":"Fed Regist"},{"key":"B40","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1016\/j.jocm.2018.07.002","article-title":"Is your dataset big enough? Sample size requirements when using artificial neural networks for discrete choice analysis","volume":"28","author":"Alwosheel","year":"2018","journal-title":"J Choice Model"},{"key":"B41","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1016\/j.jclinepi.2020.03.002","article-title":"Logistic regression was as good as machine learning for predicting major chronic diseases","volume":"122","author":"Nusinovici","year":"2020","journal-title":"J Clin Epidemiol"},{"key":"B42","doi-asserted-by":"publisher","first-page":"11222","DOI":"10.1038\/s41598-019-47712-5","article-title":"Do population-level risk prediction models that use routinely collected health data reliably predict individual risks?","volume":"9","author":"Li","year":"2019","journal-title":"Sci Rep"}],"container-title":["Frontiers in Pediatrics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fped.2023.1177470\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,6,28]],"date-time":"2023-06-28T05:35:34Z","timestamp":1687930534000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fped.2023.1177470\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,28]]},"references-count":42,"alternative-id":["10.3389\/fped.2023.1177470"],"URL":"https:\/\/doi.org\/10.3389\/fped.2023.1177470","relation":{},"ISSN":["2296-2360"],"issn-type":[{"value":"2296-2360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,6,28]]},"article-number":"1177470"}}