[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83763-en":3,"doc-seo-83763-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},83763,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","Explainable AI for Screening Abuse-Related Trauma in Bangladeshi Children: A Training-Free Multimodal Framework","Bangladesh lacks culturally adapted tools for early screening of abuse-related psychological trauma in children, with limited child mental-health specialists and no Bengali early-screening system. ShishuRaksha AI provides decision support (not diagnosis) using training-free multimodal fusion of SDQ and CPSS questionnaires, Bengali narrative text, House-Tree-Person drawing features, and facial affect. Clinically weighted cross-modal attention and perturbation-based additive attribution generate interpretable risk scores and bilingual reports with referral routing to national child-protection bodies. A noise-aware synthetic benchmark evaluates feasibility via ablations and calibration.","Explainable AI for Screening Abuse-Related Trauma in Bangladeshi Children: A Training-Free Multimodal Framework Evaluated on Noise-Aware  \nSynthetic Data  \narXiv :2607 .040 10v 1 [ cs .AI] 4 Jul 2026  \nSalma Hoque Talukdar Koli  \nDept. of Computer Science & Engineering RTM Al-Kabir Technical University Sylhet-3100, Bangladesh [info.salmahoquetalukdarkoli@gmail.com](info.salmahoquetalukdarkoli@gmail.com)  \nFahima Haque Talukder Jely  \nDept. of Computer Science & Engineering North East University Bangladesh Sylhet, Bangladesh [fahimahaquetalukderjely@gmail.com](fahimahaquetalukderjely@gmail.com)  \nAbstract—Bangladesh has an estimated 1.17 mental-health professionals per 100,000 population and only six child psychiatrists nationwide. No Bengali-language, culturally adapted tool exists for early screening of abuse-related psychological trauma in children. We present ShishuRaksha AI, a decision-support (not diagnostic) framework that fuses four screening modalities: validated questionnaires (SDQ, CPSS), Bengali narrative text, House-Tree-Person (HTP) drawing features, and facial affect. The fusion is training-free and clinically weighted, uses cross-modal attention, and includes a single-modality override rule. Every risk score is explained through clinically weighted, perturbation-based additive attribution and rendered as a bilingual (Bangla/English) report with referral routing to national child-protection services (OCC, DSS, NMHH) under the Children Act 2013. No clinical dataset of abused children can be collected ethically at this stage, so we introduce a noise-aware synthetic benchmark (500 cases, 116 positive [23.2%], four deliberate noise layers, literaturegrounded HTP priors) and evaluate tree-ensemble surrogates of the fusion design (facial channel excluded) under 5-fold stratified cross-validation. The fused model reaches an AUC of 0.874 [0.834–0.908], against 0.756 [0.705–0.803] for an SDQonly baseline, with ablation, operating-point, subgroup, and calibration analyses. We state all limitations openly, including synthetic-only data, no held-out set, text-feature circularity, and an urban–rural subgroup gap. This work is a feasibility study anda design contribution toward ethically deployable child-protection screening in low-resource settings.  \nIndex Terms—child protection, psychological trauma screening, multimodal fusion, explainable AI, Bengali NLP, synthetic benchmark, decision support, Bangladesh  \nI. INTRODUCTION  \nA child in a remote haor district such as Sunamganj who stops speaking after a violent episode at home will, in most cases, never see a specialist. Bangladesh has an estimated 1.17 mental-health professionals per 100,000 population and, per WHO figures, only six child psychiatrists in the entire country [1] . The epidemiology is not reassuring: community surveys using the Bangla SDQ have found psychiatric disorder in substantial fractions of 5–10-year-olds across rural, urban, and slum settings alike [2] . Abuse-related trauma sits in the  \nworst corner of this gap. It is stigmatized and rarely disclosed, and outside a handful of urban clinics it is almost never screened. The instruments that do exist are paper questionnaires that need trained administrators, written in English or translated without any automation.  \nMachine learning cannot close this gap, and we do not claim it can. What it might do is triage: help a school counselor, an NGO field worker, or a shishu welfare officer decide which children need referral to the scarce professionals who exist. That framing shapes every design choice here. ShishuRaksha AI 1 produces a prioritization signal with a human-readable justification, in Bangla, routed to Bangladesh’s real childprotection bodies: One-Stop Crisis Centres (OCC), the Department of Social Services (DSS), and the national mentalhealth helpline, under the Children Act 2013 [3] . It is decision support; it does not diagnose, and it is built so that it cannot quietly pretend to.  \nThere is","cbCaivb8D5Y8kzvh","https://ap.wps.com/l/cbCaivb8D5Y8kzvh","pdf",337698,4,1,6,"English","en",105,"# I. Introduction\n# II. Related Work\n# A. Child Psychological Screening Instruments","[{\"question\":\"What problem does ShishuRaksha AI target in Bangladesh?\",\"answer\":\"It targets the lack of early, culturally adapted screening for abuse-related psychological trauma in children, where most children cannot reach specialists and existing tools require trained administrators.\"},{\"question\":\"How does ShishuRaksha AI generate its risk score?\",\"answer\":\"It fuses four modalities—SDQ and CPSS questionnaires, Bengali narrative text, House-Tree-Person drawing features, and facial affect—using training-free clinically weighted cross-modal attention plus a single-modality override rule.\"},{\"question\":\"Why does the study rely on a synthetic benchmark instead of real abused-child data?\",\"answer\":\"Because collecting or validating on records of abused children cannot be ethically justified at the proof-of-concept stage, the work uses a noise-aware synthetic benchmark grounded in literature to evaluate the design with standard validation checks.\"}]",1784190279,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"explainable-ai-for-screening-abuse-related-trauma-in-bangladeshi-children-a-training-free-multimodal-framework","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/explainable-ai-for-screening-abuse-related-trauma-in-bangladeshi-children-a-training-free-multimodal-framework/83763/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does ShishuRaksha AI target in Bangladesh?","Question",{"text":75,"@type":76},"It targets the lack of early, culturally adapted screening for abuse-related psychological trauma in children, where most children cannot reach specialists and existing tools require trained administrators.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does ShishuRaksha AI generate its risk score?",{"text":80,"@type":76},"It fuses four modalities—SDQ and CPSS questionnaires, Bengali narrative text, House-Tree-Person drawing features, and facial affect—using training-free clinically weighted cross-modal attention plus a single-modality override rule.",{"name":82,"@type":73,"acceptedAnswer":83},"Why does the study rely on a synthetic benchmark instead of real abused-child data?",{"text":84,"@type":76},"Because collecting or validating on records of abused children cannot be ethically justified at the proof-of-concept stage, the work uses a noise-aware synthetic benchmark grounded in literature to evaluate the design with standard validation checks.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]