Architectures and Strategies for the Differential Diagnosis of Adolescent Depression and Anxiety: Bridging Tabular Deep Learning, Interpretability, and Scale Streamlining

Architectures and Strategies for the Differential Diagnosis of Adolescent Depression and Anxiety: Bridging Tabular Deep Learning, Interpretability, and Scale Streamlining | 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 Architectures and Strategies for the Differential Diagnosis of Adolescent Depression and Anxiety: Bridging Tabular Deep Learning, Interpretability, and Scale Streamlining Ruibing Sun, Yutian Wu, Zhongze Luo, Qizhi Zheng, Yanhong Bai, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9481972/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Background Escalating societal and occupational pressures have exacerbated the global mental health crisis. The symptomatic overlap between depression and anxiety presents a formidable diagnostic hurdle for differential diagnosis in mental health. developing AI-driven diagnostic models to differentiate these states and streamlining complex clinical instruments is of substantial clinical and scientific value. Methods Utilizing clinical questionnaires and physician-confirmed diagnoses, distinct datasets were established. This study evaluates four dimensions: model architecture, data strategy optimization, interpretability, and clinical instrument streamlining. Four resampling strategies and tabular generative models were integrated to address data imbalance. Through stratified cross-validation, 29 models—spanning traditional Machine Learning (ML), sequential Deep Learning (DL), and advanced tabular DL—were compared. Optimization strategies focusing on input representation and generalization were also evaluated. Results Data strategies significantly enhanced standard deep learning architectures. Specifically, Mambular-TTT combined with Borderline-SMOTE and TVAE achieved a 7.3% relative AUC improvement over the state-of-the-art (SOTA) baseline. Hierarchical SHAP identified childhood trauma and neuroticism as shared core risk factors while elucidating subtype-specific risk patterns. Stepwise reduction revealed that a 7-instrument subset optimally balances diagnostic performance and clinical efficiency; the Area Under the Receiver Operating Characteristic Curve (AUC) for the streamlined Dataset 2 improved to 0.70, outperforming the full-scale model. Conclusions This study demonstrates that reliable AI-assisted diagnosis requires the integrated optimization of model architecture and data strategy. This approach, combined with an interpretability-based scale streamlining strategy, provides a practical and evidence-based implementation path for transforming machine learning models into efficient, deployable clinical diagnostic tools. Differential Diagnosis of Depression and Anxiety Tabular deep learning State space models (SSM) Data balancing and synthesis Interpretability analysis Clinical instrument streamlining Full Text Additional Declarations No competing interests reported. Supplementary Files Appendix.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 13 May, 2026 Reviewers agreed at journal 13 May, 2026 Reviews received at journal 11 May, 2026 Reviewers agreed at journal 08 May, 2026 Reviewers agreed at journal 08 May, 2026 Reviewers invited by journal 08 May, 2026 Editor assigned by journal 08 May, 2026 Editor invited by journal 04 May, 2026 Submission checks completed at journal 01 May, 2026 First submitted to journal 01 May, 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-9481972","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":639543385,"identity":"22aa945d-570e-4cf1-a1cf-b319d32574d1","order_by":0,"name":"Ruibing Sun","email":"","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Ruibing","middleName":"","lastName":"Sun","suffix":""},{"id":639543386,"identity":"765365f2-c019-4a86-8238-884c865da43b","order_by":1,"name":"Yutian Wu","email":"","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Yutian","middleName":"","lastName":"Wu","suffix":""},{"id":639543387,"identity":"8a25bc05-c8b2-4c6e-8704-d81637a11301","order_by":2,"name":"Zhongze Luo","email":"","orcid":"","institution":"The Chinese University of HongKong (Shenzhen)","correspondingAuthor":false,"prefix":"","firstName":"Zhongze","middleName":"","lastName":"Luo","suffix":""},{"id":639543388,"identity":"6f52b2d0-92c0-4145-8a9b-35d92f1ed4e5","order_by":3,"name":"Qizhi Zheng","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Qizhi","middleName":"","lastName":"Zheng","suffix":""},{"id":639543389,"identity":"eb27a7d7-698b-40db-9bec-91da768d1167","order_by":4,"name":"Yanhong Bai","email":"","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Yanhong","middleName":"","lastName":"Bai","suffix":""},{"id":639543390,"identity":"5865aa5e-49b0-4d22-8f3a-f58cfcf485bc","order_by":5,"name":"Xudong Pan","email":"","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Xudong","middleName":"","lastName":"Pan","suffix":""},{"id":639543391,"identity":"f56168e1-378b-4133-b56b-7aa65773542f","order_by":6,"name":"Xiyu Ding","email":"","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Xiyu","middleName":"","lastName":"Ding","suffix":""},{"id":639543392,"identity":"0f97c983-660b-4150-a35f-8a52fef5feb7","order_by":7,"name":"Liping Liu","email":"","orcid":"","institution":"The First Psychiatric Hospital of Harbin","correspondingAuthor":false,"prefix":"","firstName":"Liping","middleName":"","lastName":"Liu","suffix":""},{"id":639543393,"identity":"d856c80d-1683-4d10-8c5f-5517f4885085","order_by":8,"name":"Guanghui Dong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYDCCAzCGBJD9gYGBH8omUsvBGQwMkg0kaWHmIUYL3/Hewy+/ttnlyUf3GB623WEjYXCA+eBtHga7PFxaJM+cS7OWbUsuNrxzxuBw7pk0oBa2ZGsehuRiXFoMbuSYGUu2MSdunJED1NJ2uM7gAI+ZNA/DgcQG/FrqIVos2/4DbeH/RkiL8cOPbYcT50sAtTC2HQBq4WHDq0XyzBkzZoZzxxM3SKQVHOxtS5aQPMxmbDnHIBmnFr7jPcYff5RVJ86fkbz5w882Owm+480Pb7ypsMOpBQhAzgC68ACMzwx2MG71ICUffwBJeTyGjoJRMApGwQgHAMmLXdlz/tJAAAAAAElFTkSuQmCC","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":true,"prefix":"","firstName":"Guanghui","middleName":"","lastName":"Dong","suffix":""},{"id":639543394,"identity":"4b9fe640-f167-4d86-92f1-b43e70befe6e","order_by":9,"name":"Xuemei Guan","email":"","orcid":"","institution":"Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Xuemei","middleName":"","lastName":"Guan","suffix":""}],"badges":[],"createdAt":"2026-04-21 09:26:50","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9481972/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9481972/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109760204,"identity":"abfd0b0b-1c2a-441f-9d40-113449061e5e","added_by":"auto","created_at":"2026-05-22 07:28:18","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6090147,"visible":true,"origin":"","legend":"","description":"","filename":"ArchitecturesandStrategiesfortheDifferentialDiagnosisofAdolescentDepressionandAnxietyBridgingTabularDeepLearningInterpretabilityandScaleStreamliningv1.2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9481972/v1_covered_f0fbdda4-29d0-4fc3-8c34-30aa3a286697.pdf"},{"id":109489238,"identity":"0c1daf5a-f883-4778-9995-e27dc1af08ae","added_by":"auto","created_at":"2026-05-18 17:12:16","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":434069,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-9481972/v1/399f3504e6c548a322c7c618.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Architectures and Strategies for the Differential Diagnosis of Adolescent Depression and Anxiety: Bridging Tabular Deep Learning, Interpretability, and Scale Streamlining","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":"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":"Differential Diagnosis of Depression and Anxiety, Tabular deep learning, State space models (SSM), Data balancing and synthesis, Interpretability analysis, Clinical instrument streamlining","lastPublishedDoi":"10.21203/rs.3.rs-9481972/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9481972/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eEscalating societal and occupational pressures have exacerbated the global mental health crisis. The symptomatic overlap between depression and anxiety presents a formidable diagnostic hurdle for differential diagnosis in mental health. developing AI-driven diagnostic models to differentiate these states and streamlining complex clinical instruments is of substantial clinical and scientific value.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eUtilizing clinical questionnaires and physician-confirmed diagnoses, distinct datasets were established. This study evaluates four dimensions: model architecture, data strategy optimization, interpretability, and clinical instrument streamlining. Four resampling strategies and tabular generative models were integrated to address data imbalance. Through stratified cross-validation, 29 models\u0026mdash;spanning traditional Machine Learning (ML), sequential Deep Learning (DL), and advanced tabular DL\u0026mdash;were compared. Optimization strategies focusing on input representation and generalization were also evaluated.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eData strategies significantly enhanced standard deep learning architectures. Specifically, Mambular-TTT combined with Borderline-SMOTE and TVAE achieved a 7.3% relative AUC improvement over the state-of-the-art (SOTA) baseline. Hierarchical SHAP identified childhood trauma and neuroticism as shared core risk factors while elucidating subtype-specific risk patterns. Stepwise reduction revealed that a 7-instrument subset optimally balances diagnostic performance and clinical efficiency; the Area Under the Receiver Operating Characteristic Curve (AUC) for the streamlined Dataset 2 improved to 0.70, outperforming the full-scale model.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThis study demonstrates that reliable AI-assisted diagnosis requires the integrated optimization of model architecture and data strategy. This approach, combined with an interpretability-based scale streamlining strategy, provides a practical and evidence-based implementation path for transforming machine learning models into efficient, deployable clinical diagnostic tools.\u003c/p\u003e","manuscriptTitle":"Architectures and Strategies for the Differential Diagnosis of Adolescent Depression and Anxiety: Bridging Tabular Deep Learning, Interpretability, and Scale Streamlining","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-18 17:12:12","doi":"10.21203/rs.3.rs-9481972/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-13T06:57:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"201915576420734293139938082439060439495","date":"2026-05-13T05:52:57+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-11T14:14:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"139664854518014681301958230768642784837","date":"2026-05-08T17:16:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"18043468165912264547596359574881642109","date":"2026-05-08T14:57:56+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-08T10:00:35+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-08T08:57:19+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-05-04T05:24:02+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-05-01T13:57:12+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Psychiatry","date":"2026-05-01T13:51:32+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":"dc881318-8a3a-4fc3-bc23-5eddcd93059f","owner":[],"postedDate":"May 18th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-13T06:57:36+00:00","index":44,"fulltext":""},{"type":"reviewerAgreed","content":"201915576420734293139938082439060439495","date":"2026-05-13T05:52:57+00:00","index":43,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-11T14:14:11+00:00","index":42,"fulltext":""},{"type":"reviewerAgreed","content":"139664854518014681301958230768642784837","date":"2026-05-08T17:16:53+00:00","index":39,"fulltext":""},{"type":"reviewerAgreed","content":"18043468165912264547596359574881642109","date":"2026-05-08T14:57:56+00:00","index":38,"fulltext":""},{"type":"reviewersInvited","content":"10","date":"2026-05-08T10:00:35+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-08T08:57:19+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-18T17:12:12+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-18 17:12:12","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9481972","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9481972","identity":"rs-9481972","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}