A Multi-Agent AI Framework for MentalHealth Triage in Post-Conflict Arabic-SpeakingPopulations

A Multi-Agent AI Framework for MentalHealth Triage in Post-Conflict Arabic-SpeakingPopulations | 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 Multi-Agent AI Framework for MentalHealth Triage in Post-Conflict Arabic-SpeakingPopulations Anas Shahin, Bahaa Masry This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9345152/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Syria’s protracted conflict has produced a mental health crisis of ex-traordinary scale, with PTSD, depression, and anxiety prevalence estimated at 7–8times global baselines, against fewer than 0.37 psychiatrists per 100,000 people. Ex-isting AI mental health tools fail in this context due to three simultaneous structuraldeficiencies—extreme clinical scarcity, Arabic NLP underperformance, and culturalmisalignment with Syrian idioms of distress—which we term the Triple Gap.Aim: To propose and specify a multi-agent AI framework for mental health triage thatjointly addresses clinical, technological, and cultural barriers in post-conflict Arabic-speaking populations, aligned with WHO’s mhGAP task-shifting model. Methods: We developed a formally specified four-agent pipeline architecture com-prising Screening, Risk Stratification, Routing, and Follow-up agents, underpinned bya cross-cutting Cultural Adaptation Layer and Human-in-the-Loop governance. Theframework was evaluated through structured comparative analysis against five exist-ing multi-agent mental health frameworks and a comprehensive four-phase validationprotocol was designed. Results: Comparative analysis demonstrates that no existing multi-agent mental healthframework addresses Arabic language support, dialect-aware NLP, cultural adapta-tion, post-conflict deployment, or task-shifting alignment. The proposed validationprotocol specifies four sequential phases—component benchmarking, simulated sce-nario testing, controlled field pilot, and stepped-wedge comparative effectiveness trial—with defined performance thresholds (symptom classification F1 ≥ 0.75; crisis sensi-tivity ≥ 95%; triage concordance κ ≥ 0.60) and ethical safeguards calibrated to fragile-state constraints. Conclusions: The proposed framework provides the first formally specified multi-agent architecture for mental health triage in Arabic-speaking, post-conflict popula-tions. The four-phase validation protocol establishes a concrete roadmap from con-ceptual design to clinical implementation, with participatory construction of a SyrianArabic mental health corpus identified as the critical prerequisite. multi-agent systems Arabic NLP mental health post-conflict task-shifting HITL governance cultural adaptation Syria mhGAP Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-9345152","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":618943645,"identity":"ab8a47c7-08e4-41fb-9638-36d6a8a78996","order_by":0,"name":"Anas Shahin","email":"data:image/png;base64,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","orcid":"","institution":"Syrian Virtual Univeristy","correspondingAuthor":true,"prefix":"","firstName":"Anas","middleName":"","lastName":"Shahin","suffix":""},{"id":618943647,"identity":"10f544b3-10fe-4e4a-8238-115822e09d77","order_by":1,"name":"Bahaa Masry","email":"","orcid":"","institution":"Damascus University","correspondingAuthor":false,"prefix":"","firstName":"Bahaa","middleName":"","lastName":"Masry","suffix":""}],"badges":[],"createdAt":"2026-04-07 12:38:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9345152/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9345152/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107870310,"identity":"f73a0ddd-c89f-400f-836d-79615b8b523e","added_by":"auto","created_at":"2026-04-27 07:39:19","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":728093,"visible":true,"origin":"","legend":"","description":"","filename":"PTSD.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9345152/v1_covered_91228591-c667-47d6-8c04-c5907299f192.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Multi-Agent AI Framework for MentalHealth Triage in Post-Conflict Arabic-SpeakingPopulations","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"multi-agent systems, Arabic NLP, mental health, post-conflict, task-shifting, HITL governance, cultural adaptation, Syria, mhGAP","lastPublishedDoi":"10.21203/rs.3.rs-9345152/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9345152/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBackground: Syria’s protracted conflict has produced a mental health crisis of ex-traordinary scale, with PTSD, depression, and anxiety prevalence estimated at 7–8times global baselines, against fewer than 0.37 psychiatrists per 100,000 people. Ex-isting AI mental health tools fail in this context due to three simultaneous structuraldeficiencies—extreme clinical scarcity, Arabic NLP underperformance, and culturalmisalignment with Syrian idioms of distress—which we term the Triple Gap.Aim: To propose and specify a multi-agent AI framework for mental health triage thatjointly addresses clinical, technological, and cultural barriers in post-conflict Arabic-speaking populations, aligned with WHO’s mhGAP task-shifting model.\u003c/p\u003e\n\u003cp\u003eMethods: We developed a formally specified four-agent pipeline architecture com-prising Screening, Risk Stratification, Routing, and Follow-up agents, underpinned bya cross-cutting Cultural Adaptation Layer and Human-in-the-Loop governance. Theframework was evaluated through structured comparative analysis against five exist-ing multi-agent mental health frameworks and a comprehensive four-phase validationprotocol was designed.\u003c/p\u003e\n\u003cp\u003eResults: Comparative analysis demonstrates that no existing multi-agent mental healthframework addresses Arabic language support, dialect-aware NLP, cultural adapta-tion, post-conflict deployment, or task-shifting alignment. The proposed validationprotocol specifies four sequential phases—component benchmarking, simulated sce-nario testing, controlled field pilot, and stepped-wedge comparative effectiveness trial—with defined performance thresholds (symptom classification F1 ≥ 0.75; crisis sensi-tivity ≥ 95%; triage concordance κ ≥ 0.60) and ethical safeguards calibrated to fragile-state constraints.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConclusions: The proposed framework provides the first formally specified multi-agent architecture for mental health triage in Arabic-speaking, post-conflict popula-tions. The four-phase validation protocol establishes a concrete roadmap from con-ceptual design to clinical implementation, with participatory construction of a SyrianArabic mental health corpus identified as the critical prerequisite.\u003c/p\u003e","manuscriptTitle":"A Multi-Agent AI Framework for MentalHealth Triage in Post-Conflict Arabic-SpeakingPopulations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-13 05:41:37","doi":"10.21203/rs.3.rs-9345152/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":"49cc8f47-f929-4260-90f6-1f08ccc4ec90","owner":[],"postedDate":"April 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-26T23:23:32+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-13 05:41:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9345152","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9345152","identity":"rs-9345152","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}