DNet :A depression recognition network combining residual network and vision transformer

DNet :A depression recognition network combining residual network and vision transformer | 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 DNet :A depression recognition network combining residual network and vision transformer Zhongyi Jiang, Xing Gao, Yin Cao, Yihan Zhang, Guanzhong Dong, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4465101/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Sep, 2025 Read the published version in BMC Psychiatry → Version 1 posted 12 You are reading this latest preprint version Abstract Depression is a prevalent and severe global mental disorder, yet its diagnosis and treatment encounter numerous challenges. This study introduces an innovative depression identification network, termed DNet. Our approach utilizes facial images and local facial images as crucial sources of data. Given that facial expressions of individuals with depression at varying severity levels inherently share similar latent facial features, subtle differences exist across multiple facial regions. To achieve higher recognition accuracy, a method is required to fuse advanced semantic features between local and global features.Therefore,we proposeDNet, comprising two key components: the Feature Extraction Module (FEM) and the Vision Transformer (ViT) Block. Specifically, FEM introduces an attention mechanism that considers both channel and positional information of the feature map. Two FEMs are employed to separately process facial and local facial images, extracting critical features to generate highly semantic information-rich feature maps. Subsequently, the feature maps of both images are concatenated along the channel dimension, and the ViT Block is utilized to comprehensively learn advanced semantic features of local and global information related to different facial expression regions. Finally, a 1×1 convolution layer and a fully connected layer are applied to adjust feature channels, yielding more robust predictive results and ultimately outputting depression prediction scores.We experimentally validate the DNet network on the AVEC2014 dataset and our self-constructed CZ2023 dataset, obtaining results of MAE=6.27, RMSE=7.96, and MAE=7.46, RMSE=9.15, respectively. These results affirm the effectiveness of the proposed method. Depression Facial Images Deep Learning ViT Dnet Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 29 Sep, 2025 Read the published version in BMC Psychiatry → Version 1 posted Editorial decision: Revision requested 11 Sep, 2024 Reviews received at journal 09 Sep, 2024 Reviews received at journal 06 Sep, 2024 Reviewers agreed at journal 01 Sep, 2024 Reviewers agreed at journal 31 Aug, 2024 Reviewers agreed at journal 15 Aug, 2024 Reviewers agreed at journal 12 Aug, 2024 Reviewers invited by journal 12 Aug, 2024 Editor invited by journal 26 Jun, 2024 Editor assigned by journal 15 Jun, 2024 Submission checks completed at journal 15 Jun, 2024 First submitted to journal 23 May, 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-4465101","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":320905519,"identity":"c7a7bf96-c45e-4db1-9081-58feec89f60f","order_by":0,"name":"Zhongyi Jiang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAt0lEQVRIiWNgGAWjYBACxgYGxgdQtgHRWphhSonUAgRsEqRpYe4/Y1bxo6YusYG9eZsEQ80dYhx2xuxmzzG2xAaeY2USDMeeEaGlscfsNmMDT2KDRI6ZBGPDYSK0NPOYFTM2SCQ2yL8hVksbjxkzY4MB0BYeYrX0sBVL9hxLMG7jSSu2SDhGhBbD/sMbPwBDTLaf/fDGGx9qiNHSwAGJDjYQkUBYAwODPAP7A2LUjYJRMApGwUgGAB7oNCU0miTLAAAAAElFTkSuQmCC","orcid":"","institution":"Changzhou University","correspondingAuthor":true,"prefix":"","firstName":"Zhongyi","middleName":"","lastName":"Jiang","suffix":""},{"id":320905521,"identity":"49d0aff3-fe5e-42e6-8cb8-2bf2c6523fd0","order_by":1,"name":"Xing Gao","email":"","orcid":"","institution":"Changzhou University","correspondingAuthor":false,"prefix":"","firstName":"Xing","middleName":"","lastName":"Gao","suffix":""},{"id":320905525,"identity":"224e8f63-65c5-414a-b661-bf526af7ccac","order_by":2,"name":"Yin Cao","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yin","middleName":"","lastName":"Cao","suffix":""},{"id":320905526,"identity":"8180f310-9e5c-4bd1-939a-e3529c99cd83","order_by":3,"name":"Yihan Zhang","email":"","orcid":"","institution":"Changzhou University","correspondingAuthor":false,"prefix":"","firstName":"Yihan","middleName":"","lastName":"Zhang","suffix":""},{"id":320905527,"identity":"9b8f93f0-1c3a-4be3-b3d8-16376f8ce673","order_by":4,"name":"Guanzhong Dong","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Guanzhong","middleName":"","lastName":"Dong","suffix":""},{"id":320905529,"identity":"7aca7db7-7116-4a33-85f4-3a4a980dc79d","order_by":5,"name":"Yun Chen","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yun","middleName":"","lastName":"Chen","suffix":""},{"id":320905531,"identity":"3747a23d-e5d0-4811-a4fc-e9ffbd4b71a5","order_by":6,"name":"Xuanyan Zhu","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xuanyan","middleName":"","lastName":"Zhu","suffix":""},{"id":320905533,"identity":"71b61140-69af-4e57-b6b6-a01d5c814004","order_by":7,"name":"Qiaoyang Zhang","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Qiaoyang","middleName":"","lastName":"Zhang","suffix":""},{"id":320905534,"identity":"4b608b2d-ce8a-4819-bb37-e6d9e0bf090f","order_by":8,"name":"Ran Bi","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ran","middleName":"","lastName":"Bi","suffix":""},{"id":320905535,"identity":"42e4e300-0bfc-47e4-ab4a-1964b02aee99","order_by":9,"name":"Ke Wang","email":"","orcid":"","institution":"Changzhou No.2 People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2024-05-23 07:42:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4465101/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4465101/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12888-025-07322-0","type":"published","date":"2025-09-29T15:57:25+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":92883933,"identity":"6c3a5e08-040e-42d8-8282-41b49c98cfa9","added_by":"auto","created_at":"2025-10-06 16:11:15","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1502997,"visible":true,"origin":"","legend":"","description":"","filename":"DNetAdepressionrecognitionnetworkcombiningresidualnetworkandvisiontransformer.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4465101/v1_covered_c5d65875-4ff8-42b1-8d55-05448d474949.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"DNet :A depression recognition network combining residual network and vision transformer","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-psychiatry","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpsy","sideBox":"Learn more about [BMC Psychiatry](http://bmcpsychiatry.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bpsy/default.aspx","title":"BMC Psychiatry","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Depression, Facial Images, Deep Learning, ViT, Dnet","lastPublishedDoi":"10.21203/rs.3.rs-4465101/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4465101/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDepression is a prevalent and severe global mental disorder, yet its diagnosis and treatment encounter numerous challenges. This study introduces an innovative depression identification network, termed DNet. Our approach utilizes facial images and local facial images as crucial sources of data. Given that facial expressions of individuals with depression at varying severity levels inherently share similar latent facial features, subtle differences exist across multiple facial regions. To achieve higher recognition accuracy, a method is required to fuse advanced semantic features between local and global features.Therefore,we proposeDNet, comprising two key components: the Feature Extraction Module (FEM) and the Vision Transformer (ViT) Block. Specifically, FEM introduces an attention mechanism that considers both channel and positional information of the feature map. Two FEMs are employed to separately process facial and local facial images, extracting critical features to generate highly semantic information-rich feature maps. Subsequently, the feature maps of both images are concatenated along the channel dimension, and the ViT Block is utilized to comprehensively learn advanced semantic features of local and global information related to different facial expression regions. Finally, a 1×1 convolution layer and a fully connected layer are applied to adjust feature channels, yielding more robust predictive results and ultimately outputting depression prediction scores.We experimentally validate the DNet network on the AVEC2014 dataset and our self-constructed CZ2023 dataset, obtaining results of MAE=6.27, RMSE=7.96, and MAE=7.46, RMSE=9.15, respectively. These results affirm the effectiveness of the proposed method.\u003c/p\u003e","manuscriptTitle":"DNet :A depression recognition network combining residual network and vision transformer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-02 06:24:15","doi":"10.21203/rs.3.rs-4465101/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-09-11T10:30:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-09T06:48:14+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-07T00:04:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"152014017794046180490843965099339162596","date":"2024-09-01T23:54:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"164501802610158063084046874984039388058","date":"2024-08-31T08:35:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"298813295515372097367072500999336794686","date":"2024-08-15T04:29:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"219919720367113803701304388507300960483","date":"2024-08-12T11:56:36+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-08-12T11:25:29+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-06-26T11:58:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-06-15T05:34:14+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-06-15T05:33:59+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Psychiatry","date":"2024-05-23T07:41:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-psychiatry","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpsy","sideBox":"Learn more about [BMC Psychiatry](http://bmcpsychiatry.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bpsy/default.aspx","title":"BMC Psychiatry","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"89fcc028-1a90-41c0-8faf-b0cc42f52492","owner":[],"postedDate":"July 2nd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-10-06T16:05:55+00:00","versionOfRecord":{"articleIdentity":"rs-4465101","link":"https://doi.org/10.1186/s12888-025-07322-0","journal":{"identity":"bmc-psychiatry","isVorOnly":false,"title":"BMC Psychiatry"},"publishedOn":"2025-09-29 15:57:25","publishedOnDateReadable":"September 29th, 2025"},"versionCreatedAt":"2024-07-02 06:24:15","video":"","vorDoi":"10.1186/s12888-025-07322-0","vorDoiUrl":"https://doi.org/10.1186/s12888-025-07322-0","workflowStages":[]},"version":"v1","identity":"rs-4465101","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4465101","identity":"rs-4465101","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}