Multimodal ICU In-hospital Mortality Prediction: A Scoping Review of Transparency, Calibration, and Translation Readiness

Multimodal ICU In-hospital Mortality Prediction: A Scoping Review of Transparency, Calibration, and Translation Readiness | 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 Systematic Review Multimodal ICU In-hospital Mortality Prediction: A Scoping Review of Transparency, Calibration, and Translation Readiness Alexander Bakumenko, Janine Hoelscher, D. Hudson Smith, Svetlana Bakumenko, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9217081/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 Background : Multimodal machine learning for intensive care unit (ICU) in-hospital mortality (IHM) prediction has rapidly expanded, yet reporting practices relevant to clinical readiness, including calibration, clinical utility, robustness to data missingness, and actionable transparency, remain heterogeneous. Objective : To map the landscape of multimodal IHM prediction studies and synthesize evidence on (i) fusion architectures and transparency implementations, (ii) evaluation beyond discrimination (calibration, clinical utility, and per-case uncertainty), and (iii) operational readiness (missing modality handling, prediction timeliness, external validation, subgroup evaluation, and reproducibility artifacts). Methods : We conducted a scoping review of peer-reviewed studies (English, January 2010 – November 2025) that fused at least two data modalities by provenance and predicted IHM in adult ICU populations. Searches were performed in PubMed, Scopus, and Google Scholar. Screening and extraction used an established AI-assisted workflow with two-reviewer verification. We summarized extracted fields from 115 included studies using descriptive statistics and narrative synthesis, including thematic analysis of author-reported gaps, limitations, and future directions. Results : Among 115 included studies, feature-level (early) fusion predominated (n = 80, 69.6%), followed by intermediate fusion (n = 30, 26.1%), late fusion (n = 4, 3.5%), and hybrid fusion (n=1, 0.9%). The modality landscape was dominated by physiologic time-series and coded/tabular EHR data; clinical text, imaging, and high-frequency waveforms each appeared in a minority of studies. Feature-level interpretability was reported in 21 studies (18.3%), with full per-case attribution in only 13 (11.3%); modality-level attribution was identified in 2 studies (1.7%), and missing-modality policies in 7 (6.1%). Calibration was assessed in 40 studies (34.8%), decision curve analysis in 20 (17.4%), and per-case uncertainty quantification in 1 (0.9%). External validation was performed in 28 studies (24.3%), prediction timeliness was evaluated in 22 (19.1%), and code repositories were provided in 29 (25.2%); no study reported a structured model card. Conclusions : The multimodal IHM literature emphasizes discrimination performance and early fusion, yet the evidence base remains immature for clinical translation, with systematic gaps in calibration, clinical utility, per-case transparency, missing-modality robustness, and reproducibility. These findings motivate standardized reporting and evaluation frameworks tailored to multimodal ICU prediction. Artificial Intelligence and Machine Learning Critical Care & Emergency Medicine ICU In-hospital mortality Multimodal learning Electronic health records Calibration Decision curve analysis Interpretability Reproducibility Scoping review Full Text Additional Declarations The authors declare no competing interests. Supplementary Files Extractiontable.pdf Extraction table 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-9217081","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":611689074,"identity":"eeebc446-5093-403f-9a6a-d580241ad19d","order_by":0,"name":"Alexander Bakumenko","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYDACCQZmECXHwA4XSiBOizGEIkVLYgPRWvhnNz825s2xSe9vZj668UfFHQZ+9hwD/JbcOWaczLstLXfGYba02zxnnjFI9rzBr4XhRoLxYd5th3M3MPOY3WZsO8xgcIOALfI30j8DtfxPN2Dm/3bzJ1CLPSEtQDNBDjuQYMDMw3aDF2SLBAEthnfOFBvO3ZZsCPSLGdAvh3kkzjwrwKtF7nb7Zom32+zk+dubn938UXFYjr89eQNeLRiAhzTlo2AUjIJRMAqwAgDFNEbmWfCG6wAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-7212-9573","institution":"Clemson University","correspondingAuthor":true,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Bakumenko","suffix":""},{"id":611689075,"identity":"6c652742-5711-418a-826b-758d8fcd9f91","order_by":1,"name":"Janine Hoelscher","email":"","orcid":"https://orcid.org/0000-0003-3970-0613","institution":"Clemson University","correspondingAuthor":false,"prefix":"","firstName":"Janine","middleName":"","lastName":"Hoelscher","suffix":""},{"id":611689076,"identity":"5cd250fa-122d-45d7-984d-7e88f5281183","order_by":2,"name":"D. Hudson Smith","email":"","orcid":"https://orcid.org/0000-0003-3041-4602","institution":"Clemson University","correspondingAuthor":false,"prefix":"","firstName":"D.","middleName":"Hudson","lastName":"Smith","suffix":""},{"id":611689077,"identity":"6299361c-c96c-4349-91f9-0c0a642585c0","order_by":3,"name":"Svetlana Bakumenko","email":"","orcid":"","institution":"Linnaeus University","correspondingAuthor":false,"prefix":"","firstName":"Svetlana","middleName":"","lastName":"Bakumenko","suffix":""},{"id":611689078,"identity":"e499738a-47ec-4ed6-91c3-b1f3daae8242","order_by":4,"name":"Aaron J. Masino","email":"","orcid":"https://orcid.org/0000-0002-2684-0548","institution":"Clemson University","correspondingAuthor":false,"prefix":"","firstName":"Aaron","middleName":"J.","lastName":"Masino","suffix":""},{"id":611689079,"identity":"d601cf62-a16d-4324-b240-4bab9286a9af","order_by":5,"name":"Jihad S. Obeid","email":"","orcid":"https://orcid.org/0000-0002-7193-7779","institution":"Medical University of South Carolina","correspondingAuthor":false,"prefix":"","firstName":"Jihad","middleName":"S.","lastName":"Obeid","suffix":""}],"badges":[],"createdAt":"2026-03-25 02:10:11","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9217081/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9217081/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105457222,"identity":"28cd0c6b-7d38-4c3e-804f-3f64724d01c3","added_by":"auto","created_at":"2026-03-26 09:21:49","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":838220,"visible":true,"origin":"","legend":"","description":"","filename":"ElsevierICUMortalitySLRRS.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9217081/v1_covered_0f313a67-f665-45ed-a7f4-2879056317d5.pdf"},{"id":105457216,"identity":"33a5d750-f284-47f4-9891-7c0eae5f2730","added_by":"auto","created_at":"2026-03-26 09:21:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":680039,"visible":true,"origin":"","legend":"\u003cp\u003eExtraction table\u003c/p\u003e","description":"","filename":"Extractiontable.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9217081/v1/165cbc17b2498a778f0ff070.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eMultimodal ICU In-hospital Mortality Prediction: A Scoping Review of Transparency, Calibration, and Translation Readiness\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Clemson University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","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":"ICU, In-hospital mortality, Multimodal learning, Electronic health records, Calibration, Decision curve analysis, Interpretability, Reproducibility, Scoping review","lastPublishedDoi":"10.21203/rs.3.rs-9217081/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9217081/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: Multimodal machine learning for intensive care unit (ICU) in-hospital mortality (IHM) prediction has rapidly expanded, yet reporting practices relevant to clinical readiness, including calibration, clinical utility, robustness to data missingness, and actionable transparency, remain heterogeneous.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective\u003c/strong\u003e: To map the landscape of multimodal IHM prediction studies and synthesize evidence on (i) fusion architectures and transparency implementations, (ii) evaluation beyond discrimination (calibration, clinical utility, and per-case uncertainty), and (iii) operational readiness (missing modality handling, prediction timeliness, external validation, subgroup evaluation, and reproducibility artifacts).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: We conducted a scoping review of peer-reviewed studies (English, January 2010 – November 2025) that fused at least two data modalities by provenance and predicted IHM in adult ICU populations. Searches were performed in PubMed, Scopus, and Google Scholar. Screening and extraction used an established AI-assisted workflow with two-reviewer verification. We summarized extracted fields from 115 included studies using descriptive statistics and narrative synthesis, including thematic analysis of author-reported gaps, limitations, and future directions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Among 115 included studies, feature-level (early) fusion predominated (n = 80, 69.6%), followed by intermediate fusion (n = 30, 26.1%), late fusion (n = 4, 3.5%), and hybrid fusion (n=1, 0.9%). The modality landscape was dominated by physiologic time-series and coded/tabular EHR data; clinical text, imaging, and high-frequency waveforms each appeared in a minority of studies. Feature-level interpretability was reported in 21 studies (18.3%), with full per-case attribution in only 13 (11.3%); modality-level attribution was identified in 2 studies (1.7%), and missing-modality policies in 7 (6.1%). Calibration was assessed in 40 studies (34.8%), decision curve analysis in 20 (17.4%), and per-case uncertainty quantification in 1 (0.9%). External validation was performed in 28 studies (24.3%), prediction timeliness was evaluated in 22 (19.1%), and code repositories were provided in 29 (25.2%); no study reported a structured model card.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e: The multimodal IHM literature emphasizes discrimination performance and early fusion, yet the evidence base remains immature for clinical translation, with systematic gaps in calibration, clinical utility, per-case transparency, missing-modality robustness, and reproducibility. These findings motivate standardized reporting and evaluation frameworks tailored to multimodal ICU prediction.\u003c/p\u003e","manuscriptTitle":"Multimodal ICU In-hospital Mortality Prediction: A Scoping Review of Transparency, Calibration, and Translation Readiness","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-26 09:21:31","doi":"10.21203/rs.3.rs-9217081/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","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":"9aa5ede9-45f8-4a3e-826b-9c3d8dd2d379","owner":[],"postedDate":"March 26th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65082937,"name":"Artificial Intelligence and Machine Learning"},{"id":65082938,"name":"Critical Care \u0026 Emergency Medicine"}],"tags":[],"updatedAt":"2026-03-26T09:21:31+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-26 09:21:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9217081","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9217081","identity":"rs-9217081","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}