A Novel ViT-BILSTM Model for Physical Activity Intensity classification in Adults using Gravity-based Acceleration

A Novel ViT-BILSTM Model for Physical Activity Intensity classification in Adults using Gravity-based Acceleration | 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 Novel ViT-BILSTM Model for Physical Activity Intensity classification in Adults using Gravity-based Acceleration Lin Wang, Zizhang Luo, Tianle Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4696057/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 01 Feb, 2025 Read the published version in BMC Biomedical Engineering → Version 1 posted 4 You are reading this latest preprint version Abstract Aim The aim of this study is to apply a novel hybrid framework incorporating a Vision Transformer (ViT) and bidirectional long short-term memory (Bi-LSTM) model for classifying physical activity intensity (PAI) in adults using gravity-based acceleration. Additionally, it further investigates how PAI and temporal window (TW) impacts the model’ s accuracy. Method This research used the Capture-24 dataset, consisting of raw accelerometer data from 151 participants aged 18 to 91. Gravity-based acceleration was utilised to generate images encoding various PAIs. These images were subsequently analysed using the ViT-BiLSTM model, with results presented in confusion matrices and compared with baseline models. The model's robustness was evaluated through temporal stability testing and examination of accuracy and loss curves. Result The ViT-BiLSTM model excelled in PAI classification task, achieving an overall accuracy of 98.5% ±1.48% across five TWs-98.7% for 1s, 98.1% for 5s, 98.2% for 10s, 99% for 15s, and 98.65% for 30s of TW. The model consistently exhibited superior accuracy in predicting sedentary (98.9%±1%) compared to light physical activity (98.2%±2%) and moderate-to-vigorous physical activity (98.2%± 3%). ANOVA showed no significant accuracy variation across PAIs (F = 2.18, p = 0.13) and TW (F = 0.52, p = 0.72). Accuracy and loss curves show the model consistently improves its performance across epochs, demonstrating its excellent robustness. Conclusion This study demonstrates the ViT-BiLSTM model’s efficacy in classifying PAI using gravity-based acceleration, with performance remaining consistent across diverse TWs and intensities. However, PAI and TW could result in slight variations in the model’s performance. Future research should concern and investigate the impact of gravity-based acceleration on PAI thresholds, which may influence model's robustness and reliability. Deep learning Raw accelerometer data Variation Generalisation. Physical activity patterns Full Text Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterials1Modelresults.xlsx Supplementarymaterials2ConfusionMatrices.docx Supplementarymaterials3AccuracyandLossCurvesfordifferentTWs.docx Cite Share Download PDF Status: Published Journal Publication published 01 Feb, 2025 Read the published version in BMC Biomedical Engineering → Version 1 posted Editorial decision: Revision requested 16 Jul, 2024 Editor assigned by journal 16 Jul, 2024 Submission checks completed at journal 12 Jul, 2024 First submitted to journal 06 Jul, 2024 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-4696057","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":327501805,"identity":"2b549c8f-c5c5-4ad8-908e-87c70978757c","order_by":0,"name":"Lin Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0ElEQVRIiWNgGAWjYBACPgbm4z8+VCAL8RDQwsbAliA54wyYRbQWHgVp3jaStEjkMBjzzrNJ7JdvPvaBocaOweDMAUJacg8kzt2WZizZxpY8g+FYMoPB2QYCWqTzEg683XZYzuAYjzGQe4DB4Dwhh0nnGDbwzvnPY3+M/zMDwz/itBgz8jYckDNg42FmYGw7QITD5J+lMc44lmwscSzNmCGxL5lHkpD3+XkOH2P4UGOX2N98+DHDh292cnxnEgi4DAUkEI6VUTAKRsEoGAXEAABPWju5HOc5LwAAAABJRU5ErkJggg==","orcid":"","institution":"University of Exeter","correspondingAuthor":true,"prefix":"","firstName":"Lin","middleName":"","lastName":"Wang","suffix":""},{"id":327501809,"identity":"ebac68d0-e546-4c46-a136-f4f0577df20f","order_by":1,"name":"Zizhang Luo","email":"","orcid":"","institution":"Yangtze University","correspondingAuthor":false,"prefix":"","firstName":"Zizhang","middleName":"","lastName":"Luo","suffix":""},{"id":327501810,"identity":"00d7fc9a-e71d-41a4-8d45-173a6d771d9c","order_by":2,"name":"Tianle Zhang","email":"","orcid":"","institution":"University of Liverpool","correspondingAuthor":false,"prefix":"","firstName":"Tianle","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2024-07-06 09:23:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4696057/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4696057/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s42490-025-00088-2","type":"published","date":"2025-02-01T15:57:26+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":75351305,"identity":"32199bcb-ea3a-422c-bdf8-3b1ee4fbc6a6","added_by":"auto","created_at":"2025-02-03 16:09:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":837148,"visible":true,"origin":"","legend":"","description":"","filename":"ImpactofEpochLengthonVisionANovelViTBILSTMModelforPhysicalActivityIntensityinAdultsusingGravitybasedAcceleration0907.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4696057/v1_covered_31a1863e-9677-45cb-9266-54ad800b5bb5.pdf"},{"id":61892125,"identity":"0e256288-6ae6-42ed-a063-dc4ae0dad059","added_by":"auto","created_at":"2024-08-06 18:41:16","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":16177,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterials1Modelresults.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4696057/v1/616ffed686884671b8f7173f.xlsx"},{"id":61892126,"identity":"eb3b4042-577c-4dc6-9837-7bdd29f2c6f2","added_by":"auto","created_at":"2024-08-06 18:41:16","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":2150697,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterials2ConfusionMatrices.docx","url":"https://assets-eu.researchsquare.com/files/rs-4696057/v1/f26033e06606ffff28652dc6.docx"},{"id":61892128,"identity":"63bfe4b0-2747-4029-96c0-f958b408086c","added_by":"auto","created_at":"2024-08-06 18:41:16","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":229041,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterials3AccuracyandLossCurvesfordifferentTWs.docx","url":"https://assets-eu.researchsquare.com/files/rs-4696057/v1/47e4c7d039a6eebced56668d.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Novel ViT-BILSTM Model for Physical Activity Intensity classification in Adults using Gravity-based Acceleration","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-biomedical-engineering","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bbme","sideBox":"Learn more about [BMC Biomedical Engineering](http://bmcbiomedeng.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bbme/default.aspx","title":"BMC Biomedical Engineering","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Deep learning, Raw accelerometer data, Variation, Generalisation. Physical activity patterns","lastPublishedDoi":"10.21203/rs.3.rs-4696057/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4696057/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eAim\u003c/h2\u003e \u003cp\u003eThe aim of this study is to apply a novel hybrid framework incorporating a Vision Transformer (ViT) and bidirectional long short-term memory (Bi-LSTM) model for classifying physical activity intensity (PAI) in adults using gravity-based acceleration. Additionally, it further investigates how PAI and temporal window (TW) impacts the model\u0026rsquo; s accuracy.\u003c/p\u003e\u003ch2\u003eMethod\u003c/h2\u003e \u003cp\u003eThis research used the Capture-24 dataset, consisting of raw accelerometer data from 151 participants aged 18 to 91. Gravity-based acceleration was utilised to generate images encoding various PAIs. These images were subsequently analysed using the ViT-BiLSTM model, with results presented in confusion matrices and compared with baseline models. The model's robustness was evaluated through temporal stability testing and examination of accuracy and loss curves.\u003c/p\u003e\u003ch2\u003eResult\u003c/h2\u003e \u003cp\u003eThe ViT-BiLSTM model excelled in PAI classification task, achieving an overall accuracy of 98.5% \u0026plusmn;1.48% across five TWs-98.7% for 1s, 98.1% for 5s, 98.2% for 10s, 99% for 15s, and 98.65% for 30s of TW. The model consistently exhibited superior accuracy in predicting sedentary (98.9%\u0026plusmn;1%) compared to light physical activity (98.2%\u0026plusmn;2%) and moderate-to-vigorous physical activity (98.2%\u0026plusmn; 3%). ANOVA showed no significant accuracy variation across PAIs (F\u0026thinsp;=\u0026thinsp;2.18, p\u0026thinsp;=\u0026thinsp;0.13) and TW (F\u0026thinsp;=\u0026thinsp;0.52, p\u0026thinsp;=\u0026thinsp;0.72). Accuracy and loss curves show the model consistently improves its performance across epochs, demonstrating its excellent robustness.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis study demonstrates the ViT-BiLSTM model\u0026rsquo;s efficacy in classifying PAI using gravity-based acceleration, with performance remaining consistent across diverse TWs and intensities. However, PAI and TW could result in slight variations in the model\u0026rsquo;s performance. Future research should concern and investigate the impact of gravity-based acceleration on PAI thresholds, which may influence model's robustness and reliability.\u003c/p\u003e","manuscriptTitle":"A Novel ViT-BILSTM Model for Physical Activity Intensity classification in Adults using Gravity-based Acceleration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-06 18:41:11","doi":"10.21203/rs.3.rs-4696057/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-07-16T06:17:45+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-16T05:24:54+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-07-12T06:05:53+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Biomedical Engineering","date":"2024-07-06T09:21:10+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-biomedical-engineering","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bbme","sideBox":"Learn more about [BMC Biomedical Engineering](http://bmcbiomedeng.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bbme/default.aspx","title":"BMC Biomedical Engineering","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"316aa6a9-71d8-4975-923c-22a730c36956","owner":[],"postedDate":"August 6th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-02-03T16:02:05+00:00","versionOfRecord":{"articleIdentity":"rs-4696057","link":"https://doi.org/10.1186/s42490-025-00088-2","journal":{"identity":"bmc-biomedical-engineering","isVorOnly":false,"title":"BMC Biomedical Engineering"},"publishedOn":"2025-02-01 15:57:26","publishedOnDateReadable":"February 1st, 2025"},"versionCreatedAt":"2024-08-06 18:41:11","video":"","vorDoi":"10.1186/s42490-025-00088-2","vorDoiUrl":"https://doi.org/10.1186/s42490-025-00088-2","workflowStages":[]},"version":"v1","identity":"rs-4696057","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4696057","identity":"rs-4696057","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}