Convolutional neural networks improved HRV analysis accuracy by single-lead Holter

Convolutional neural networks improved HRV analysis accuracy by single-lead Holter | 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 Convolutional neural networks improved HRV analysis accuracy by single-lead Holter Chunping Tang, Qiong Huang, Liang Yuan, Ningtian Zhou This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2709337/v2 This work is licensed under a CC BY 4.0 License Status: Posted Version 2 posted You are reading this latest preprint version Show more versions Abstract Objective: New artificial intelligence (AI) algorithms are being applied to HRV but there is still needed for more comparison with classical HRV metrics. Convolutional Neural Network (CNN) was used to analyze HRV in four different groups distinguished by body mass index (BMI) and age. Methods: The cohort study enrolled total 265 patients wore an AI single-lead Holter and traditional multi-lead Holter for less than 22 h from March 1, 2023, to December 1, 2024. Indeed, RR-interval sequence as input for the CNN, then linear fitting and Bland–Altman analysis were used to illustrate the statistical results of AI Holter and traditional Holter in four groups: BMI <24 kg/m 2 and age <65 years, BMI <24 kg/m 2 and age ≥65 years, BMI ≥24 kg/m 2 and age <65 years, and BMI ≥24 kg/m 2 and age ≥65 years. Results: All groups had acceptable biases and r-values for different HRV parameters. SDANN was the most accurate HRV parameter in all groups, and SDNN, PNN50 also showed better test efficiency in specific groups. Conclusions: The AI single-lead Holter was reliable for HRV detection, and SDANN showed a satisfactory accuracy in all groups, but SDNN and PNN50 showed better test efficiency in specific groups. Heart rate variability (HRV) Holter Artificial intelligence Aging Body mass index (BMI) Figures Figure 1 Figure 2 Figure 3 Figure 4 Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 2 posted You are reading this latest preprint version Show more versions 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-2709337","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":186005130,"identity":"8a050a11-f8c2-454a-8864-569cc089fa20","order_by":0,"name":"Chunping Tang","email":"","orcid":"","institution":"the First Affiliated Hospital of Nanjing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Chunping","middleName":"","lastName":"Tang","suffix":""},{"id":186005131,"identity":"05bad4b1-3636-41d5-919b-b38c579d4056","order_by":1,"name":"Qiong Huang","email":"","orcid":"","institution":"the First Affiliated Hospital of Nanjing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Qiong","middleName":"","lastName":"Huang","suffix":""},{"id":186005132,"identity":"3050b9a6-09ae-4931-96b6-eb17a69da57d","order_by":2,"name":"Liang Yuan","email":"","orcid":"","institution":"the First Affiliated Hospital of Nanjing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Liang","middleName":"","lastName":"Yuan","suffix":""},{"id":186005135,"identity":"46fcd27f-2b5d-47cc-82a8-282b177e1656","order_by":3,"name":"Ningtian Zhou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAElEQVRIie3SwWqDMBjA8U8+0Mv3AJb4EIGCXdnQV2kJ9CRSKIydpNCrsKvQl9hp1zkC9iHWW6HnDsZwsMO+zJWeUnssLH+IRvBHiBHA5brC0spcJQ/kcaBufjbvSAjNw+WEI7MQwQUExaoR8/k2eQlwv7uLinwE2Lx/QJLbiB81M1HJvSrRHw0z0ovx0lfrCNTCRijMYkFSK95LLDKqp081DTGEerq0kPBEgk9xQ0U/kX8kIaRY8Hf7Jd7hLJmpWyYT0nQ/KHkvUvsKQSorSSv1+kbfOg0eN8/hV1nkcrPSXvuQWMmx7gWvhIn5DZB6T4dX624tE0PbfuFyuVz/px+UP0sqJeFt9gAAAABJRU5ErkJggg==","orcid":"","institution":"the First Affiliated Hospital of Nanjing Medical University","correspondingAuthor":true,"prefix":"","firstName":"Ningtian","middleName":"","lastName":"Zhou","suffix":""}],"badges":[],"createdAt":"2023-03-19 02:59:06","currentVersionCode":2,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-2709337/v2","doiUrl":"https://doi.org/10.21203/rs.3.rs-2709337/v2","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87609443,"identity":"5fefa702-fee3-4158-a261-636878bb0fc8","added_by":"auto","created_at":"2025-07-25 19:55:09","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":389146,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIllustrations of the CNN. \u003c/strong\u003eA. The convolution process. B. The overall architecture of a CNN.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-2709337/v2/7ecb6352caf432a2af037f2c.png"},{"id":87609445,"identity":"05ec22ac-657a-4373-9097-249668fc73ff","added_by":"auto","created_at":"2025-07-25 19:55:09","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1490150,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe training and results of AI single-lead Holter. \u003c/strong\u003eA. The recording of different arrhythmias by AI single-lead Holter. B. Power Spectral Density analysis by welch, AR, and Lomb-scargle methods. C. ECG scattergram based on inter-beat interval (IBI) detection.D. Atrial fibrillation (AF) training by SWK database and MIT-BIH database (red-dots: AF, black-dots: normal beats). E. Total training of ventricular arrhythmia and supraventricular arrhythmia (blue-dots: ventricular arrhythmia, red-dots: supraventricular arrhythmia).\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-2709337/v2/0538c0372a2d29c7a40f7207.png"},{"id":87609446,"identity":"587f5e52-0931-4c72-acc7-d40bb64911cf","added_by":"auto","created_at":"2025-07-25 19:55:09","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1195638,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBland\u003c/strong\u003e–\u003cstrong\u003eAltman Analysis and 95% Limits of Agreement.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-2709337/v2/19e2d8efa15d63d94f3796b9.png"},{"id":87609444,"identity":"420b49ce-d8aa-4d8a-8dd8-3894c8ef8176","added_by":"auto","created_at":"2025-07-25 19:55:09","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1035698,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLinear Correlation Analysis.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-2709337/v2/9804f1b2fc20c45a35dea2f8.png"},{"id":87609957,"identity":"2eb4cc93-fd80-4cbe-863f-f0275d2bc69e","added_by":"auto","created_at":"2025-07-25 20:11:13","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1270296,"visible":true,"origin":"","legend":"","description":"","filename":"maintext.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2709337/v2_covered_b448a061-0e59-4c5c-94d9-9111672be136.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"Convolutional neural networks improved HRV analysis accuracy by single-lead Holter","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":"Heart rate variability (HRV), Holter, Artificial intelligence, Aging, Body mass index (BMI)","lastPublishedDoi":"10.21203/rs.3.rs-2709337/v2","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2709337/v2","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective: \u003c/strong\u003eNew artificial intelligence (AI) algorithms are being applied to HRV but there is still needed for more comparison with classical HRV metrics. Convolutional Neural Network (CNN) was used to analyze HRV in four different groups distinguished by body mass index (BMI) and age.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e The cohort study enrolled total 265 patients wore an AI single-lead Holter and traditional multi-lead Holter for less than 22 h from March 1, 2023, to December 1, 2024. Indeed, RR-interval sequence as input for the CNN, then linear fitting and Bland–Altman analysis were used to illustrate the statistical results of AI Holter and traditional Holter in four groups: BMI \u0026lt;24 kg/m\u003csup\u003e2 \u003c/sup\u003eand age \u0026lt;65 years, BMI \u0026lt;24 kg/m\u003csup\u003e2\u003c/sup\u003e and age ≥65 years, BMI ≥24 kg/m\u003csup\u003e2\u003c/sup\u003e and age \u0026lt;65 years, and BMI ≥24 kg/m\u003csup\u003e2\u003c/sup\u003e and age ≥65 years.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e All groups had acceptable biases and r-values for different HRV parameters.\u003cstrong\u003e \u003c/strong\u003eSDANN was the most accurate HRV parameter in all groups, and SDNN, PNN50 also showed better test efficiency in specific groups.\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e The AI single-lead Holter was reliable for HRV detection, and SDANN showed a satisfactory accuracy in all groups, but SDNN and PNN50 showed better test efficiency in specific groups.\u003c/p\u003e","manuscriptTitle":"Convolutional neural networks improved HRV analysis accuracy by single-lead Holter","msid":"","msnumber":"","nonDraftVersions":[{"code":2,"date":"2025-07-25 19:55:04","doi":"10.21203/rs.3.rs-2709337/v2","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}},{"code":1,"date":"2023-03-23 17:59:52","doi":"10.21203/rs.3.rs-2709337/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":"d4ebbca7-0aa5-4d15-8f95-e38e79aea780","owner":[],"postedDate":"July 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-01-17T19:30:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-25 19:55:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v2","identity":"rs-2709337","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2709337","identity":"rs-2709337","version":["v2"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}