{"paper_id":"24dcb951-2000-4aaa-9833-77a8e4b357a4","body_text":"Cardiovascular and Autonomic Phenotypes Reveal Distinct Mechanisms of Sepsis Decompensation via Deep Learning | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Cardiovascular and Autonomic Phenotypes Reveal Distinct Mechanisms of Sepsis Decompensation via Deep Learning Tilendra Choudhary, Haoming Shi, Ayman Ali, Victor Moas, Omer Inan, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9136766/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Sepsis heterogeneity reflects diverse etiologies and patient-specific physiological responses, motivating phenotype identification to enable precision therapeutics. However, most phenotyping approaches rely on intermittently sampled clinical variables, whereas continuously recorded physiological waveforms remain underutilized. We developed a deep-learning framework to derive physiological phenotypes from five-minute pre-onset electrocardiogram, photoplethysmogram and respiratory-impedance waveforms in 2,174 ICU patients meeting Sepsis-3 criteria. From these signals, 192 cardiorespiratory physiomarkers were extracted and embedded using a Feature Tokenizer Transformer encoder, which outperformed alternative representation methods. Consensus clustering identified four stable sepsis physio-phenotypes (SP-1–SP-4) associated with distinct autonomic and peripheral vascular signatures. Despite similar baseline severity and demographics, phenotypes differed significantly in mortality (19–29%), septic shock, vasopressor use and mechanical ventilation, with divergent 28-day survival trajectories (P<0.01). Explainable AI provided clinically interpretable characterizations, and a trained classifier enabled real-time bedside phenotyping. This framework establishes waveform-based phenotyping as a foundation for precision medicine in sepsis care. Health sciences/Biomarkers/Prognostic markers Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Computational biology and bioinformatics/Computational models Health sciences/Diseases/Infectious diseases/Bacterial infection Physical sciences/Engineering/Biomedical engineering Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementalDocumentFinal.pdf Supplemental Document Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-9136766\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":608379499,\"identity\":\"4dc02127-d1dc-4810-82b6-9941893732d4\",\"order_by\":0,\"name\":\"Tilendra Choudhary\",\"email\":\"data:image/png;base64,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\",\"orcid\":\"\",\"institution\":\"Duke University\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Tilendra\",\"middleName\":\"\",\"lastName\":\"Choudhary\",\"suffix\":\"\"},{\"id\":608379500,\"identity\":\"6adc2f51-25a3-4d88-b796-73e8bc28a398\",\"order_by\":1,\"name\":\"Haoming Shi\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Duke University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Haoming\",\"middleName\":\"\",\"lastName\":\"Shi\",\"suffix\":\"\"},{\"id\":608379501,\"identity\":\"3faba9f3-8fcb-40b1-a111-ca54daa3ae4b\",\"order_by\":2,\"name\":\"Ayman Ali\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-2791-2793\",\"institution\":\"\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Ayman\",\"middleName\":\"\",\"lastName\":\"Ali\",\"suffix\":\"\"},{\"id\":608379502,\"identity\":\"c0dc7de0-616a-4e56-a7b5-2b51508d558d\",\"order_by\":3,\"name\":\"Victor Moas\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-9341-0428\",\"institution\":\"Duke University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Victor\",\"middleName\":\"\",\"lastName\":\"Moas\",\"suffix\":\"\"},{\"id\":608379503,\"identity\":\"c17c17b7-d514-4a90-9a71-cb535b80d476\",\"order_by\":4,\"name\":\"Omer Inan\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Georgia Institute of Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Omer\",\"middleName\":\"\",\"lastName\":\"Inan\",\"suffix\":\"\"},{\"id\":608379504,\"identity\":\"7fc2c580-14f4-4969-bfb0-a9f67beb237a\",\"order_by\":5,\"name\":\"Mihai Podgoreanu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Duke University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Mihai\",\"middleName\":\"\",\"lastName\":\"Podgoreanu\",\"suffix\":\"\"},{\"id\":608379505,\"identity\":\"58276338-5db3-4d1e-9511-3c3176a2f0ff\",\"order_by\":6,\"name\":\"Vijay Krishnamoorthy\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Duke University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Vijay\",\"middleName\":\"\",\"lastName\":\"Krishnamoorthy\",\"suffix\":\"\"},{\"id\":608379506,\"identity\":\"ca0d99a5-dfda-4759-ac26-ab903ab0bc08\",\"order_by\":7,\"name\":\"Craig Jabaley\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-6687-3953\",\"institution\":\"Emory University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Craig\",\"middleName\":\"\",\"lastName\":\"Jabaley\",\"suffix\":\"\"},{\"id\":608379507,\"identity\":\"7350ea70-7b71-4309-bacc-a06ff168ff3b\",\"order_by\":8,\"name\":\"Craig Coopersmith\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-2400-3217\",\"institution\":\"Emory University School of Medicine\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Craig\",\"middleName\":\"\",\"lastName\":\"Coopersmith\",\"suffix\":\"\"},{\"id\":608379508,\"identity\":\"0672d46a-a6bf-4398-ac1e-8fef110ef293\",\"order_by\":9,\"name\":\"Michael Pinsky\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-6166-700X\",\"institution\":\"University of Pittsburgh\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Michael\",\"middleName\":\"\",\"lastName\":\"Pinsky\",\"suffix\":\"\"},{\"id\":608379509,\"identity\":\"d5123bfa-a1ae-4cd7-88bf-084f9fdfb1fd\",\"order_by\":10,\"name\":\"Gilles Clermont\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-0163-1379\",\"institution\":\"University of Pittsburgh\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Gilles\",\"middleName\":\"\",\"lastName\":\"Clermont\",\"suffix\":\"\"},{\"id\":608379510,\"identity\":\"150c30eb-8ae7-4b89-915d-29caddb68547\",\"order_by\":11,\"name\":\"Rishikesan Kamaleswaran\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Duke University School of Medicine\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Rishikesan\",\"middleName\":\"\",\"lastName\":\"Kamaleswaran\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2026-03-16 10:46:26\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-9136766/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-9136766/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":105563664,\"identity\":\"191dad6f-d3c0-4763-9748-719ca7ea23a8\",\"added_by\":\"auto\",\"created_at\":\"2026-03-27 12:47:25\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1851116,\"visible\":true,\"origin\":\"\",\"legend\":\"Article File\",\"description\":\"\",\"filename\":\"ManuscriptFinal.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9136766/v1_covered_8f8d5189-af09-4e27-974b-11c12f52bd2c.pdf\"},{\"id\":105171775,\"identity\":\"f877912c-302a-42ab-8d17-fa7e55d3366f\",\"added_by\":\"auto\",\"created_at\":\"2026-03-23 04:10:00\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":4699929,\"visible\":true,\"origin\":\"\",\"legend\":\"Supplemental Document\",\"description\":\"\",\"filename\":\"SupplementalDocumentFinal.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9136766/v1/d901959c9b312c20f311a2bc.pdf\"}],\"financialInterests\":\"There is \\u003cb\\u003eNO\\u003c/b\\u003e Competing Interest.\",\"formattedTitle\":\"Cardiovascular and Autonomic Phenotypes Reveal Distinct Mechanisms of Sepsis Decompensation via Deep Learning\",\"fulltext\":[],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":false,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":true,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":true,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-9136766/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-9136766/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"Sepsis heterogeneity reflects diverse etiologies and patient-specific physiological responses, motivating phenotype identification to enable precision therapeutics. However, most phenotyping approaches rely on intermittently sampled clinical variables, whereas continuously recorded physiological waveforms remain underutilized. We developed a deep-learning framework to derive physiological phenotypes from five-minute pre-onset electrocardiogram, photoplethysmogram and respiratory-impedance waveforms in 2,174 ICU patients meeting Sepsis-3 criteria. From these signals, 192 cardiorespiratory physiomarkers were extracted and embedded using a Feature Tokenizer Transformer encoder, which outperformed alternative representation methods. Consensus clustering identified four stable sepsis physio-phenotypes (SP-1–SP-4) associated with distinct autonomic and peripheral vascular signatures. Despite similar baseline severity and demographics, phenotypes differed significantly in mortality (19–29%), septic shock, vasopressor use and mechanical ventilation, with divergent 28-day survival trajectories (P\\u003c0.01). Explainable AI provided clinically interpretable characterizations, and a trained classifier enabled real-time bedside phenotyping. This framework establishes waveform-based phenotyping as a foundation for precision medicine in sepsis care.\",\"manuscriptTitle\":\"Cardiovascular and Autonomic Phenotypes Reveal Distinct Mechanisms of Sepsis Decompensation via Deep Learning\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2026-03-23 04:09:47\",\"doi\":\"10.21203/rs.3.rs-9136766/v1\",\"editorialEvents\":[],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"cbe03136-7c28-48a9-a1a8-894f6774c80a\",\"owner\":[],\"postedDate\":\"March 23rd, 2026\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[{\"id\":64737924,\"name\":\"Health sciences/Biomarkers/Prognostic markers\"},{\"id\":64737925,\"name\":\"Biological sciences/Computational biology and bioinformatics/Machine learning\"},{\"id\":64737926,\"name\":\"Biological sciences/Computational biology and bioinformatics/Computational models\"},{\"id\":64737927,\"name\":\"Health sciences/Diseases/Infectious diseases/Bacterial infection\"},{\"id\":64737928,\"name\":\"Physical sciences/Engineering/Biomedical engineering\"}],\"tags\":[],\"updatedAt\":\"2026-03-23T04:09:48+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2026-03-23 04:09:47\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-9136766\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-9136766\",\"identity\":\"rs-9136766\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}