Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Care

Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Care | 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 Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Care Yuxiao Cheng, Xinxin Song, Ziqian Wang, Qin Zhong, Qionghai Dai, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5773165/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 Recent advances in deep learning (DL) have prompted the development of high-performing early warning score (EWS) systems, predicting clinical deteriorations such as acute kidney injury, acute myocardial infarction, or circulatory failure. DL models have proven to be powerful tools for various tasks but come with the cost of lacking interpretability and limited generalizability, hindering their clinical applications. To develop a practical EWS system applicable to various outcomes, we propose causally-informed explainable early prediction model, which leverages causal discovery to identify the underlying causal relationships of prediction and thus owns two unique advantages: demonstrating the explicit interpretation of the prediction while exhibiting decent performance when applied to unfamiliar environments. Benefiting from these features, our approach achieves superior accuracy for 6 different critical deteriorations and achieves better generalizability across different patient groups, compared to various baseline algorithms. Besides, we provide explicit causal pathways to serve as references for assistant clinical diagnosis and potential interventions. The proposed approach enhances the practical application of deep learning in various medical scenarios. Health sciences/Health care/Prognosis Health sciences/Medical research/Outcomes research Full Text Additional Declarations There is NO Competing Interest. Supplementary Files cdeepsupp.pdf Supplementary materials 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-5773165","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":398319516,"identity":"99c5a1b8-7168-4891-a682-2f6e9158adea","order_by":0,"name":"Yuxiao Cheng","email":"data:image/png;base64,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","orcid":"","institution":"Tsinghua University","correspondingAuthor":true,"prefix":"","firstName":"Yuxiao","middleName":"","lastName":"Cheng","suffix":""},{"id":398319517,"identity":"8e121065-63b2-4291-a9d1-6b392838b617","order_by":1,"name":"Xinxin Song","email":"","orcid":"","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Xinxin","middleName":"","lastName":"Song","suffix":""},{"id":398319518,"identity":"aaa4c7a2-d83e-4206-94b9-a9b95c47ac6c","order_by":2,"name":"Ziqian Wang","email":"","orcid":"","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Ziqian","middleName":"","lastName":"Wang","suffix":""},{"id":398319519,"identity":"89cdb307-d201-40be-b53e-c342fb6d6af7","order_by":3,"name":"Qin Zhong","email":"","orcid":"","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Qin","middleName":"","lastName":"Zhong","suffix":""},{"id":398319520,"identity":"b5c4f024-a546-404d-bb1a-1fb85c5b2107","order_by":4,"name":"Qionghai Dai","email":"","orcid":"https://orcid.org/0000-0001-7043-3061","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Qionghai","middleName":"","lastName":"Dai","suffix":""},{"id":398319521,"identity":"9cf28fce-088c-4d0b-83eb-5924f4af94d1","order_by":5,"name":"Kunlun He","email":"","orcid":"https://orcid.org/0000-0002-3335-5700","institution":"Chinese PLA General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Kunlun","middleName":"","lastName":"He","suffix":""},{"id":398319522,"identity":"14ba3e65-4775-43dd-999b-b3e8d487ec00","order_by":6,"name":"Jinli Suo","email":"","orcid":"https://orcid.org/0000-0002-3426-1634","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Jinli","middleName":"","lastName":"Suo","suffix":""}],"badges":[],"createdAt":"2025-01-06 10:59:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5773165/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5773165/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":79460790,"identity":"c801a143-788a-47e8-90d7-7058eb0dd9e5","added_by":"auto","created_at":"2025-03-28 17:09:54","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2174302,"visible":true,"origin":"","legend":"Article File","description":"","filename":"cdeepmanuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5773165/v1_covered_23131039-3d96-46ab-a2d4-5662faf7a2b2.pdf"},{"id":78868303,"identity":"96a6075f-0c67-4317-812f-af96db8c29fe","added_by":"auto","created_at":"2025-03-20 05:02:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4398080,"visible":true,"origin":"","legend":"Supplementary materials","description":"","filename":"cdeepsupp.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5773165/v1/20f42ebb7845b58f6a30303f.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Care","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":"[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":"","lastPublishedDoi":"10.21203/rs.3.rs-5773165/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5773165/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Recent advances in deep learning (DL) have prompted the development of high-performing early warning score (EWS) systems, predicting clinical deteriorations such as acute kidney injury, acute myocardial infarction, or circulatory failure. DL models have proven to be powerful tools for various tasks but come with the cost of lacking interpretability and limited generalizability, hindering their clinical applications. To develop a practical EWS system applicable to various outcomes, we propose causally-informed explainable early prediction model, which leverages causal discovery to identify the underlying causal relationships of prediction and thus owns two unique advantages: demonstrating the explicit interpretation of the prediction while exhibiting decent performance when applied to unfamiliar environments. Benefiting from these features, our approach achieves superior accuracy for 6 different critical deteriorations and achieves better generalizability across different patient groups, compared to various baseline algorithms. Besides, we provide explicit causal pathways to serve as references for assistant clinical diagnosis and potential interventions. The proposed approach enhances the practical application of deep learning in various medical scenarios.","manuscriptTitle":"Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Care","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-20 05:02:01","doi":"10.21203/rs.3.rs-5773165/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":"86c92e04-bce4-4bfb-9277-ed63e3d07c78","owner":[],"postedDate":"March 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":42431455,"name":"Health sciences/Health care/Prognosis"},{"id":42431456,"name":"Health sciences/Medical research/Outcomes research"}],"tags":[],"updatedAt":"2025-03-28T17:01:44+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-20 05:02:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5773165","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5773165","identity":"rs-5773165","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}