{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T11:01:51Z","timestamp":1771930911453,"version":"3.50.1"},"reference-count":24,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T00:00:00Z","timestamp":1770595200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Blunt thoracic aortic injury (BTAI) is rare but highly lethal. Despite advances such as thoracic endovascular aortic repair (TEVAR), management remains challenging due to heterogeneous physiologic presentations. Injury grade alone may not capture early derangements influencing risk. In other acute syndromes, machine-learning\u2013derived physiologic phenotypes have informed prognosis. We applied this approach to BTAI to identify phenotypes, examine outcomes, and use explainable AI (XAI) to characterize their physiologic drivers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      We analyzed 1,375 patients from the Aortic Trauma Foundation registry using eleven admission physiologic and injury severity variables. After plausibility checks and median imputation, K-means clustering (\n                      <jats:italic>k<\/jats:italic>\n                      \u2009=\u20093; silhouette\u2009=\u20090.23) defined physiologic phenotypes. Patients missing &amp;gt;50% physiologic data were excluded; the remaining 1,083 were assigned to the nearest phenotype centroid. Outcomes included in-hospital mortality, treatment modality, hospital and ICU length of stay (LOS), ventilator days, and time to repair. For XAI, an XGBoost classifier was trained to reproduce K-means labels, and SHapley Additive exPlanations (SHAP) quantified feature importance and stability.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Three phenotypes emerged: \u201cStable\u201d (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009431; preserved hemodynamics, low lactate), \u201cShock\u201d (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009398; hypotension, tachycardia, high lactate), and \u201cNeurologically Compromised\u201d (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009254; depressed Glasgow Coma Score with intermediate hemodynamics). Mortality was 7.2%, 18.9%, and 18.1% (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&amp;lt;\u20090.001), significantly higher in both high-risk groups vs. Stable. TEVAR predominated (58.2\u201363.0% across phenotypes); open repair was uncommon (\u22646.3%), and medical management was more frequent in Stable (31.1% vs. Shock 19.1%;\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.0010). ICU LOS differed overall (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.008), with Neurologically Compromised requiring longer stays. The XGBoost surrogate reproduced K-means phenotypes with high fidelity (accuracy 0.917; macro-F1 0.906;\n                      <jats:italic>\u03ba<\/jats:italic>\n                      0.873; adjusted Rand index 0.788). SHAP identified systolic blood pressure, lactate, and heart rate as dominant phenotype-defining features, with stable rankings across 200 bootstrap refits.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Machine learning identified three physiologic BTAI phenotypes with distinct presentations and mortality despite similar management patterns. XAI showed that perfusion and metabolic markers, not anatomy alone, drive phenotype structure. These data reinforce the potential for physiologic phenotyping to enhance prognostication and guide BTAI decision-making.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1745987","type":"journal-article","created":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T07:46:29Z","timestamp":1770623189000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Physiologic phenotypes in blunt thoracic aortic injury: implications for risk stratification and surgical decision-making using machine learning"],"prefix":"10.3389","volume":"8","author":[{"given":"Phillip D.","family":"Jenkins","sequence":"first","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael R.","family":"Kolesnikov","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shelby","family":"Willis","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Victor","family":"Andujo","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tiffany","family":"Lian","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ruchi","family":"Thanawala","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Castigliano","family":"Bhamidipati","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Justin","family":"Regner","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Julie","family":"Doberne","sequence":"additional","affiliation":[{"name":"Department of Surgery, Oregon Health & Science University","place":["Portland, OR, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1016\/j.ejvs.2021.09.043","article-title":"Blunt traumatic aortic injury management, a French trauma base analytic cohort","volume":"63","author":"Boutin","year":"2022","journal-title":"Eur J Vasc Endovasc Surg"},{"key":"B2","doi-asserted-by":"publisher","first-page":"62","DOI":"10.1186\/s13019-020-01101-6","article-title":"Blunt thoracic aortic injury\u2014concepts and management","volume":"15","author":"Mouawad","year":"2020","journal-title":"J Cardiothorac Surg"},{"key":"B3","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1097\/TA.0b013e3181df68b3","article-title":"Blunt thoracic aortic injuries: an autopsy study","volume":"70","author":"Teixeira","year":"2011","journal-title":"J Trauma"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1006","DOI":"10.1097\/TA.0000000000001555","article-title":"The evolution of care improves outcome in blunt thoracic aortic injury: a western trauma association multicenter study","volume":"83","author":"Shackford","year":"2017","journal-title":"J Trauma Acute Care Surg"},{"key":"B5","doi-asserted-by":"publisher","first-page":"422","DOI":"10.1016\/j.avsg.2023.09.076","article-title":"Development and validation of a risk prediction tool for in-hospital mortality after thoracic endovascular repair in patients with blunt thoracic aortic injury using the aortic trauma foundation registry","volume":"99","author":"D'Oria","year":"2024","journal-title":"Ann Vasc Surg"},{"key":"B6","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1097\/TA.0000000000004322","article-title":"Using machine learning to predict outcomes of patients with blunt traumatic aortic injuries","volume":"97","author":"Lu","year":"2024","journal-title":"J Trauma Acute Care Surg"},{"key":"B7","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1016\/j.jvs.2023.03.021","article-title":"Outcomes following thoracic endovascular aortic repair for blunt thoracic aortic injury stratified by society for vascular surgery grade","volume":"78","author":"Yadavalli","year":"2023","journal-title":"J Vasc Surg"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1961","DOI":"10.1007\/s00068-020-01432-y","article-title":"Blunt thoracic aortic injury and TEVAR: long-term outcomes and health-related quality of life","volume":"48","author":"Hundersmarck","year":"2022","journal-title":"Eur J Trauma Emerg Surg"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1624","DOI":"10.1016\/j.jvs.2025.07.035","article-title":"Outcomes of routine surveillance of grade 1 blunt traumatic aortic injury","volume":"82","author":"Zaidi","year":"2025","journal-title":"J Vasc Surg"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1580","DOI":"10.1016\/j.jvs.2016.05.070","article-title":"Nonoperative management of grade III blunt thoracic aortic injuries","volume":"64","author":"Gandhi","year":"2016","journal-title":"J Vasc Surg"},{"key":"B11","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1186\/s13019-022-02094-0","article-title":"Traumatic blunt thoracic aortic injury: a 10-year single-center retrospective analysis","volume":"17","author":"Sun","year":"2022","journal-title":"J Cardiothorac Surg"},{"key":"B12","doi-asserted-by":"publisher","first-page":"965","DOI":"10.1016\/S2213-2600(23)00237-0","article-title":"Identifying molecular phenotypes in sepsis: an analysis of two prospective observational cohorts and secondary analysis of two randomised controlled trials","volume":"11","author":"Sinha","year":"2023","journal-title":"Lancet Respir Med"},{"key":"B13","doi-asserted-by":"publisher","first-page":"2136","DOI":"10.1007\/s00134-020-06296-9","article-title":"Phenotypes and personalized medicine in the acute respiratory distress syndrome","volume":"46","author":"Matthay","year":"2020","journal-title":"Intensive Care Med"},{"key":"B14","doi-asserted-by":"publisher","first-page":"228","DOI":"10.1186\/s13054-022-04079-w","article-title":"Clustering identifies endotypes of traumatic brain injury in an intensive care cohort: a CENTER-TBI study","volume":"26","author":"\u00c5kerlund","year":"2022","journal-title":"Crit Care (London, England)"},{"key":"B15","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1186\/s13054-020-2778-x","article-title":"ARDS subphenotypes: understanding a heterogeneous syndrome","volume":"24","author":"Wilson","year":"2020","journal-title":"Crit Care"},{"key":"B16","doi-asserted-by":"publisher","first-page":"e48297","DOI":"10.2196\/48297","article-title":"Machine learning\u2013enabled clinical information systems using fast healthcare interoperability resources data standards: scoping review","volume":"11","author":"Balch","year":"2023","journal-title":"JMIR Med Inform"},{"key":"B17","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1016\/j.jvs.2017.10.084","article-title":"Delayed endovascular aortic repair is associated with reduced in-hospital mortality in patients with blunt thoracic aortic injury","volume":"68","author":"Marcaccio","year":"2018","journal-title":"J Vasc Surg"},{"key":"B18","doi-asserted-by":"publisher","first-page":"651","DOI":"10.1016\/j.patrec.2009.09.011","article-title":"Data clustering: 50 years beyond K-means","volume":"31","author":"Jain","year":"2010","journal-title":"Pattern Recognit Lett"},{"key":"B19","doi-asserted-by":"publisher","first-page":"166","DOI":"10.1186\/s40168-020-00934-6","article-title":"VALENCIA: a nearest centroid classification method for vaginal microbial communities based on composition","volume":"8","author":"France","year":"2020","journal-title":"Microbiome"},{"key":"B20","doi-asserted-by":"publisher","first-page":"453","DOI":"10.1038\/s41746-025-01806-9","article-title":"Identifying clusters of people with multiple long-term conditions using large language models: a population-based study","volume":"8","author":"Smith","year":"2025","journal-title":"npj Digit Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From local explanations to global understanding with explainable AI for trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nat Mach Intell"},{"key":"B22","doi-asserted-by":"publisher","first-page":"999","DOI":"10.1097\/TA.0b013e3182858a3e","article-title":"Lactate clearance as a predictor of mortality in trauma patients","volume":"74","author":"Odom","year":"2013","journal-title":"J Trauma Acute Care Surg"},{"key":"B23","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1186\/s12873-020-00383-2","article-title":"Equivalency between the shock index and subtracting the systolic blood pressure from the heart rate: an observational cohort study","volume":"20","author":"Kamikawa","year":"2020","journal-title":"BMC Emerg Med"},{"key":"B24","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1016\/j.avsg.2025.09.023","article-title":"The role of gender in blunt thoracic aortic injuries","volume":"123","author":"D'Alessio","year":"2026","journal-title":"Ann Vasc Surg"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1745987\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T10:00:05Z","timestamp":1771927205000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1745987\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,9]]},"references-count":24,"alternative-id":["10.3389\/fdgth.2026.1745987"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1745987","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,9]]},"article-number":"1745987"}}