ICU Mortality and LOS Prediction Models Using MachineLearning Based on Both Real and Simulated Data

ICU Mortality and LOS Prediction Models Using MachineLearning Based on Both Real and Simulated Data | 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 ICU Mortality and LOS Prediction Models Using MachineLearning Based on Both Real and Simulated Data Girma Neshir Alemneh, Hirut Bekele Ashagrie, Lemlem Kassa Tegegne This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8503522/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 Health institutions in low-resource settings have limited and skewed clinical data, whichcomplicates mortality prediction and resource utilization. This shortcomings areaddressed by introducing machine learning models that are trained on the combineddata from the actual and synthetic patient populations to predict ICU patients’mortality and Length of Stay (LOS). Taking a dataset of 10,810 patient records fromfive hospitals in Ethiopia, we evaluated three machine learning algorithms-LogisticRegression (LG), Random Forest and XGBoost- across three data settings: real-only,synthetic-only (by taking SMOTE-NC as an example) and mixed configurations of realversus synthetic data. Our findings show that hybrid models perform better, with thebest-performing hybrid models achieving a mean absolute error (MAE) of approximately5.5 days for LOS prediction and XGBoost achieving 99.5% accuracy for mortalityprediction. The ICU patient features such as age, pulse rate, oxygen saturation, andhemoglobin levels are important indicators of ICU outcomes in Ethiopia.A prototypewas created to show model performance and offer useful information. This studyprovides a solid foundation for strategically integrating synthetic data to improvepredictive analytics in healthcare settings with limited resources. Biological sciences/Computational biology and bioinformatics Health sciences/Health care Physical sciences/Mathematics and computing Health sciences/Medical research Full Text Additional Declarations No competing interests reported. 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-8503522","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":573809978,"identity":"b9cbb805-8218-46a8-8e61-c6b810aa5c79","order_by":0,"name":"Girma Neshir Alemneh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEElEQVRIiWNgGAWjYDACZhhDgrGB4QOMw0OsFsYZMNU8hLRBtAC1w1Xi0yLvzvt0ww+GO3Lys5vbPtvU2NnbSyQwPnjbxiBnj0OL4WF2s5s9DM+MDe4cbJ6dcyw5sUcigdlwbhuDMS5bDJvZ2G7wMBxO3CCR2Mycw8acwCORwCbN28aQ2INHy80/DIfr588AarH4V28P1ML+G6ilHpcWeWY2tttAWxIYbgC1MLYdZgQ6jI0ZqCUBl8MMQFpkDA4bbgBqYeztO57Yc+Zhs+SccxKGPQdw2NJ/jO3mm4rD8vIz0h8z/PhWbc/ennzww5syG3n2Bhy2gI0yQBFjBKmVwOEsoC04jBoFo2AUjIJRgAAA3sBQ3n1Wp5MAAAAASUVORK5CYII=","orcid":"","institution":"Addis Ababa Science and Technology University","correspondingAuthor":true,"prefix":"","firstName":"Girma","middleName":"Neshir","lastName":"Alemneh","suffix":""},{"id":573809980,"identity":"d9e4cf75-090b-4443-9ed9-942b4adb493f","order_by":1,"name":"Hirut Bekele Ashagrie","email":"","orcid":"","institution":"St. Paul’s Hospital Millennium Medical College","correspondingAuthor":false,"prefix":"","firstName":"Hirut","middleName":"Bekele","lastName":"Ashagrie","suffix":""},{"id":573809981,"identity":"7499b4ed-89df-48bf-8976-0eecabf22528","order_by":2,"name":"Lemlem Kassa Tegegne","email":"","orcid":"","institution":"Addis Ababa Science and Technology University","correspondingAuthor":false,"prefix":"","firstName":"Lemlem","middleName":"Kassa","lastName":"Tegegne","suffix":""}],"badges":[],"createdAt":"2026-01-02 22:38:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8503522/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8503522/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100199906,"identity":"efd1dd44-5fd2-4157-b96e-e8df482bf311","added_by":"auto","created_at":"2026-01-14 04:29:12","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5357,"visible":true,"origin":"","legend":"","description":"","filename":"cd431f16943e416eb2d8a7bc6d4b78b9.json","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/d113971e86de7a9f6f4b5fc5.json"},{"id":100199910,"identity":"70eaf26e-4408-49b2-8af8-8e5d14e08556","added_by":"auto","created_at":"2026-01-14 04:29:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":263693,"visible":true,"origin":"","legend":"","description":"","filename":"ArticlesEditorcheckRevised.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/98adfdf7d02c53256488695b.pdf"},{"id":100199911,"identity":"a58dee37-9eee-4ced-bbf6-c2d651e31256","added_by":"auto","created_at":"2026-01-14 04:29:13","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":145129,"visible":true,"origin":"","legend":"","description":"","filename":"EthicalClearancemerged.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/a1feacab6105258bb4c1da29.pdf"},{"id":100199909,"identity":"0f3ba287-e1b7-409e-99e2-da945c001a4b","added_by":"auto","created_at":"2026-01-14 04:29:12","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":263693,"visible":true,"origin":"","legend":"","description":"","filename":"LaTeXManuscriptEditorRevised.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/9c9c7dde8c700422e39751dc.pdf"},{"id":100199912,"identity":"016f3edb-f89c-425e-a551-6bdbe125d742","added_by":"auto","created_at":"2026-01-14 04:29:13","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":163522,"visible":true,"origin":"","legend":"","description":"","filename":"ResponsetoEditorCheckV2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/77f571c35fb865408d170742.pdf"},{"id":100199908,"identity":"4c78c21c-2c75-48f6-9c1e-dcb9a6d72051","added_by":"auto","created_at":"2026-01-14 04:29:12","extension":"bst","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38071,"visible":true,"origin":"","legend":"","description":"","filename":"snapp2015.bst","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/80751e11040e760b5937e8ff.bst"},{"id":100199907,"identity":"230edcd5-ef49-4e58-afc0-446e4b9fb359","added_by":"auto","created_at":"2026-01-14 04:29:12","extension":"xml","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":59690,"visible":true,"origin":"","legend":"","description":"","filename":"cd431f16943e416eb2d8a7bc6d4b78b91structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1/e9024b846f136f71b271107a.xml"},{"id":108577279,"identity":"404e9c8a-7d76-4c23-b39c-5ee608088cbb","added_by":"auto","created_at":"2026-05-06 07:26:13","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":184902,"visible":true,"origin":"","legend":"","description":"","filename":"LaTeXManuscriptEditorRevised.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8503522/v1_covered_485d7d86-b1df-4e2e-8d02-bfff9588fce0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"ICU Mortality and LOS Prediction Models Using MachineLearning Based on Both Real and Simulated Data","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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-8503522/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8503522/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Health institutions in low-resource settings have limited and skewed clinical data, whichcomplicates mortality prediction and resource utilization. This shortcomings areaddressed by introducing machine learning models that are trained on the combineddata from the actual and synthetic patient populations to predict ICU patients’mortality and Length of Stay (LOS). Taking a dataset of 10,810 patient records fromfive hospitals in Ethiopia, we evaluated three machine learning algorithms-LogisticRegression (LG), Random Forest and XGBoost- across three data settings: real-only,synthetic-only (by taking SMOTE-NC as an example) and mixed configurations of realversus synthetic data. Our findings show that hybrid models perform better, with thebest-performing hybrid models achieving a mean absolute error (MAE) of approximately5.5 days for LOS prediction and XGBoost achieving 99.5% accuracy for mortalityprediction. The ICU patient features such as age, pulse rate, oxygen saturation, andhemoglobin levels are important indicators of ICU outcomes in Ethiopia.A prototypewas created to show model performance and offer useful information. This studyprovides a solid foundation for strategically integrating synthetic data to improvepredictive analytics in healthcare settings with limited resources.","manuscriptTitle":"ICU Mortality and LOS Prediction Models Using MachineLearning Based on Both Real and Simulated Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-14 04:29:08","doi":"10.21203/rs.3.rs-8503522/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":"1ef5af44-a5eb-4e05-997b-f1dcaecf3033","owner":[],"postedDate":"January 14th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":61051876,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":61051877,"name":"Health sciences/Health care"},{"id":61051878,"name":"Physical sciences/Mathematics and computing"},{"id":61051879,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-06T07:25:42+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-14 04:29:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8503522","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8503522","identity":"rs-8503522","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}