A Three-stage model for hospital length of Stay prediction using Machine and deep learning Algorithms

A Three-stage model for hospital length of Stay prediction using Machine and deep learning Algorithms | 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 Research Article A Three-stage model for hospital length of Stay prediction using Machine and deep learning Algorithms Somnath Ghosh, Debotosh Bhattacharjee This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6523042/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 The rising cost of hospitalization is a significant concern for the general public. Hospital costs are due to various diseases and their severity, leading to differing lengths of hospital stays. This paper presents a predictive analysis of hospital stay duration, accurately predicting the length of stay manage hospital operations(cost) more precisely. This study employs a dual predictive analysis method, examining overall hospital length of stay and specific disease levels. The analysis used the 2017 de-identified dataset from the USA SPARCS, the all-disease study involving 2.3 million data points, while the disease-level prediction was performed on a subset of 995,000 records. The disease-level analysis focuses on predicting and analyzing the length of stay for twenty-one diagnoses. The dual predictive analysis is structured around a three-stage hospital inpatient model, which includes the admission, post-admission, and discharge stages. The methodology employed in this study is the weighted probability ensemble model, representing a novel approach to predicting lengths of stay in the context of multiple diseases. For all diseases, the performance of the ensemble model is good. Among the three stages, the discharge stage model shows the highest performance, followed by the post-admission stage, and the admission stage performs the least. The precision, recall, and F1-score achieved in these stages are as follows: 0.44, 0.82, and 0.85 for discharge, 0.26, 0.78, and 0.82 for post-admission, and 0.27, 0.80, and 0.84 for admission, respectively. The diseases showing the best predictive performance include schizophrenia and other psychotic disorders, mood disorders, live births, HIV infection, and alcohol-related disorders. Conversely, acute myocardial infarction and osteoarthritis ranked among the lowest in performance among the selected diseases. Healthcare Hospitalization Length of stay Generic model Disease-specific model Ensemble learning 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-6523042","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":450644172,"identity":"62fbf1b0-4919-46ad-b814-a35d88f04a40","order_by":0,"name":"Somnath Ghosh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABC0lEQVRIiWNgGAWjYDACCQYGZiiTjflPhYQciHXgAXFamNkYeM5YGIO1JBCthbetIrEBxManRT66+fDngj820ebs5489kGyTSJ8fdvgh0BY7Od0G7FoM7xxLk57Zlpa7syeZ3cDgnETuxttpBkAtycZmB3BomZFjxszbcDh3w4FkNomEMqCW2QkgLQcSt+HWYvyZ5w9Qy/nHbBIH2CTSDWenf8CrRV4ix0Cahw2o5UYym2RDm0SCvHQOflsMJNLSpHmBftlw47GZNMMZCcMN0jkFBxIMcPtFfkbyYaDDbIAOS3wmzVBRJy8/O33zhw8VdnK4tBhgiENEDLArB9vSQFhkFIyCUTAKRjoAAFuHYtLx8g5HAAAAAElFTkSuQmCC","orcid":"","institution":"Jadavpur University","correspondingAuthor":true,"prefix":"","firstName":"Somnath","middleName":"","lastName":"Ghosh","suffix":""},{"id":450644173,"identity":"9b93a9ed-d79d-420f-bc71-1b9a90fcee10","order_by":1,"name":"Debotosh Bhattacharjee","email":"","orcid":"","institution":"Jadavpur University","correspondingAuthor":false,"prefix":"","firstName":"Debotosh","middleName":"","lastName":"Bhattacharjee","suffix":""}],"badges":[],"createdAt":"2025-04-24 18:08:05","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6523042/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6523042/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87952846,"identity":"3c20d839-9f5c-4a60-829e-ebd0469660ed","added_by":"auto","created_at":"2025-07-30 18:01:40","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2131704,"visible":true,"origin":"","legend":"","description":"","filename":"LospredictionjournalpaperSpringerm.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6523042/v1_covered_08af1ceb-afc1-4525-b8eb-324565c3f9c8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Three-stage model for hospital length of Stay prediction using Machine and deep learning Algorithms","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":"Healthcare, Hospitalization, Length of stay, Generic model, Disease-specific model, Ensemble learning","lastPublishedDoi":"10.21203/rs.3.rs-6523042/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6523042/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe rising cost of hospitalization is a significant concern for the general public. Hospital costs are due to various diseases and their severity, leading to differing lengths of hospital stays. This paper presents a predictive analysis of hospital stay duration, accurately predicting the length of stay manage hospital operations(cost) more precisely. This study employs a dual predictive analysis method, examining overall hospital length of stay and specific disease levels. The analysis used the 2017 de-identified dataset from the USA SPARCS, the all-disease study involving 2.3\u0026nbsp;million data points, while the disease-level prediction was performed on a subset of 995,000 records. The disease-level analysis focuses on predicting and analyzing the length of stay for twenty-one diagnoses. The dual predictive analysis is structured around a three-stage hospital inpatient model, which includes the admission, post-admission, and discharge stages. The methodology employed in this study is the weighted probability ensemble model, representing a novel approach to predicting lengths of stay in the context of multiple diseases. For all diseases, the performance of the ensemble model is good. Among the three stages, the discharge stage model shows the highest performance, followed by the post-admission stage, and the admission stage performs the least. The precision, recall, and F1-score achieved in these stages are as follows: 0.44, 0.82, and 0.85 for discharge, 0.26, 0.78, and 0.82 for post-admission, and 0.27, 0.80, and 0.84 for admission, respectively. The diseases showing the best predictive performance include schizophrenia and other psychotic disorders, mood disorders, live births, HIV infection, and alcohol-related disorders. Conversely, acute myocardial infarction and osteoarthritis ranked among the lowest in performance among the selected diseases.\u003c/p\u003e","manuscriptTitle":"A Three-stage model for hospital length of Stay prediction using Machine and deep learning Algorithms","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-05 05:29:26","doi":"10.21203/rs.3.rs-6523042/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":"549952bf-c10e-4d3b-a06c-907a367bf863","owner":[],"postedDate":"May 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-07-30T17:53:29+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-05 05:29:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6523042","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6523042","identity":"rs-6523042","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}