An unsupervised decision-support framework for multivariate biomarker analysis in athlete monitoring

An unsupervised decision-support framework for multivariate biomarker analysis in athlete monitoring | 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 An unsupervised decision-support framework for multivariate biomarker analysis in athlete monitoring Fernando Barcelos Rosito, Sebastião De Jesus Menezes, Simone Ferreira Sturza, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8996167/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Purpose. Athlete monitoring in practice is constrained by small cohorts, heterogeneous biomarker scales, limited feasibility of repeated biological sampling, and the lack of reliable injury ground truth. These limitations reduce the interpretability and utility of traditional univariate and binary risk models. This study aims to address these challenges by proposing an unsupervised multivariate framework to identify latent physiological states in athletes using real data. Methods. We propose a modular computational framework that operates directly in the joint biomarker space and integrates data preprocessing with clinical safety screening, unsupervised clustering, and centroid-based physiological interpretation. Physiological profiles are learned exclusively from data collected from amateur soccer players during a competitive microcycle. Synthetic data augmentation is applied to evaluate robustness and scalability. Ward hierarchical clustering is used for monitoring and etiological differentiation , while Gaussian Mixture Model (GMM) augmentation supports structural stability analysis in high-dimensional settings. Results. The framework identifies physiologically coherent profiles that distinguish mechanical damage from metabolic stress while preserving the dominance of homeostatic states. Results from synthetic data augmentation demonstrate the framework’s feasibility and ability to detect latent, silent risk phenotypes that are typically missed by conventional univariate monitoring. Structural stability analyses indicate that the framework remains robust under data augmentation and higher-dimensional settings. Conclusion. The proposed framework enables interpretable identification of latent physiological states from multivariate biomarker data without reliance on injury labels. By distinguishing underlying physiological mechanisms and revealing silent risk patterns not captured by conventional monitoring, it provides actionable insights to support clinicians, physiologists, and sports health professionals in individualized athlete monitoring and informed decision making. Health sciences/Biomarkers Biological sciences/Computational biology and bioinformatics Health sciences/Health care Physical sciences/Mathematics and computing Artificial Intelligence Biomarkers Workload Monitoring Cluster Analysis Gaussian Mixture Models Healthcare Informatics Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 22 Apr, 2026 Reviewers agreed at journal 22 Apr, 2026 Reviewers invited by journal 22 Apr, 2026 Editor assigned by journal 20 Apr, 2026 Editor invited by journal 09 Mar, 2026 Submission checks completed at journal 05 Mar, 2026 First submitted to journal 05 Mar, 2026 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-8996167","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":628173343,"identity":"dbfe9673-3634-44b4-b46c-8f43d408be9e","order_by":0,"name":"Fernando Barcelos Rosito","email":"","orcid":"","institution":"Universidade Federal de Ciências da Saúde de Porto Alegre","correspondingAuthor":false,"prefix":"","firstName":"Fernando","middleName":"Barcelos","lastName":"Rosito","suffix":""},{"id":628173344,"identity":"339730df-2c80-4253-8881-1a3a31df9e06","order_by":1,"name":"Sebastião De Jesus Menezes","email":"","orcid":"","institution":"Levino Inova","correspondingAuthor":false,"prefix":"","firstName":"Sebastião","middleName":"De Jesus","lastName":"Menezes","suffix":""},{"id":628173345,"identity":"a1eb4e7b-f7e3-4eae-a8f6-adabb124f783","order_by":2,"name":"Simone Ferreira Sturza","email":"","orcid":"","institution":"Levino Inova","correspondingAuthor":false,"prefix":"","firstName":"Simone","middleName":"Ferreira","lastName":"Sturza","suffix":""},{"id":628173346,"identity":"84d31817-49d9-4e7e-a4a3-81c7cf8fd4b8","order_by":3,"name":"Adriana Seixas","email":"","orcid":"","institution":"Universidade Federal de Ciências da Saúde de Porto Alegre","correspondingAuthor":false,"prefix":"","firstName":"Adriana","middleName":"","lastName":"Seixas","suffix":""},{"id":628173347,"identity":"6de0a9ff-a545-4e94-b89f-7338f717068c","order_by":4,"name":"Muriel Figueredo Franco","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDUlEQVRIie3RP0vDQBgG8CcEnCJZ39D4EYSTQBXqh2koJEstjhmk3BQXpWsE8WN0bjhIl7hn1MWpw7l1EPFNtYJypLoJ3rPcM9yP9/4ANjZ/NAucw3+v2bbABaiLCARyU+tt2UHwSZz8B+SwOHtcaAE6vrwv9fOdIn/GBdkglr3rBxPpN6koCyZhPRkFN3NF1ExGhDqNZbgUZpJAeQJTwrjv7s/VVJAnyMlVLCkxHmxDXngK+Ssmt4qEX0dr53UHQUuonSKZ8DhyZAepn1BeCRa0ioKiStu7JCfDKo3ysDKTZeLqdXbKBxsfaX0xaF9MNVwOZr3cSD7y5Qs8YMjLXhf4Fu8Xe21sbGz+Q94AuvFaFwpGJrMAAAAASUVORK5CYII=","orcid":"","institution":"Universidade Federal de Ciências da Saúde de Porto Alegre","correspondingAuthor":true,"prefix":"","firstName":"Muriel","middleName":"Figueredo","lastName":"Franco","suffix":""}],"badges":[],"createdAt":"2026-02-28 14:39:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8996167/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8996167/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107686129,"identity":"50c3210c-4af3-4eb5-aac5-ec37f049c7f8","added_by":"auto","created_at":"2026-04-24 04:24:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":532874,"visible":true,"origin":"","legend":"","description":"","filename":"JournalBiomarkersCorrelation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8996167/v1_covered_fc841546-9843-4d93-aeb1-e87761312e76.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"An unsupervised decision-support framework for multivariate biomarker analysis in athlete monitoring","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Artificial Intelligence, Biomarkers, Workload Monitoring, Cluster Analysis, Gaussian Mixture Models, Healthcare Informatics","lastPublishedDoi":"10.21203/rs.3.rs-8996167/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8996167/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose.\u003c/strong\u003e Athlete monitoring in practice is constrained by small cohorts, heterogeneous biomarker scales, limited feasibility of repeated biological sampling, and the lack of reliable injury ground truth. These limitations reduce the interpretability and utility of traditional univariate and binary risk models. This study aims to address these challenges by proposing an unsupervised multivariate framework to identify latent physiological states in athletes using real data.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods.\u003c/strong\u003e We propose a modular computational framework that operates directly in the joint biomarker space and integrates data preprocessing with clinical safety screening, unsupervised clustering, and centroid-based physiological interpretation. Physiological profiles are learned exclusively from data collected from amateur soccer players during a competitive microcycle. Synthetic data augmentation is applied to evaluate robustness and scalability. Ward hierarchical clustering is used for monitoring and etiological differentiation , while Gaussian Mixture Model (GMM) augmentation supports structural stability analysis in high-dimensional settings.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults.\u003c/strong\u003e The framework identifies physiologically coherent profiles that distinguish mechanical damage from metabolic stress while preserving the dominance of homeostatic states. Results from synthetic data augmentation demonstrate the framework’s feasibility and ability to detect latent, silent risk phenotypes that are typically missed by conventional univariate monitoring. Structural stability analyses indicate that the framework remains robust under data augmentation and higher-dimensional settings.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion.\u003c/strong\u003e The proposed framework enables interpretable identification of latent physiological states from multivariate biomarker data without reliance on injury labels. By distinguishing underlying physiological mechanisms and revealing silent risk patterns not captured by conventional monitoring, it provides actionable insights to support clinicians, physiologists, and sports health professionals in individualized athlete monitoring and informed decision making.\u003c/p\u003e","manuscriptTitle":"An unsupervised decision-support framework for multivariate biomarker analysis in athlete monitoring","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-24 04:24:02","doi":"10.21203/rs.3.rs-8996167/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-22T16:23:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"254638365435360592242703545449031931145","date":"2026-04-22T13:51:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-22T10:25:10+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-20T15:59:11+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-09T17:21:59+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-05T22:29:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-05T16:22:58+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"535e8287-1bda-4ab4-86bf-5b8bd43da6fb","owner":[],"postedDate":"April 24th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":66859952,"name":"Health sciences/Biomarkers"},{"id":66859953,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":66859954,"name":"Health sciences/Health care"},{"id":66859955,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-04-24T04:24:02+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-24 04:24:02","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8996167","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8996167","identity":"rs-8996167","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}