[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83863-en":3,"doc-seo-83863-105":29,"detail-sidebar-cat-0-en-105":90},{"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":11,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83863,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",8,"Research & Report","Medi-Gemma Hybrid Clinical Decision Support System Integrating Deterministic EMR Analytics and Retrieval-Augmented Generation","Large Language Models in clinical environments face structural hallucinations, weak deterministic reasoning on tabular patient data, and omissions from vector-based retrieval. Medi-Gemma presents a hybrid Clinical Decision Support System for wound pathology triage and workflow automation. A centralized ClinicalOrchestrator coordinates a multi-stage pipeline that cleans EMR data deterministically, routes queries via intent routing, and uses a Ground Truth Injection Module to override prompts with the latest validated patient snapshot. ProtocolManager and SafetyVerifier enforce evidence-based risk pathways and prevent unsafe output rules. Validation shows reduced context drift, fewer backend data crashes, and improved factual adherence to clinical repositories.","Medi-Gemma: A Hybrid Clinical Decision Support System Integrating Deterministic EMR Analytics and Retrieval-Augmented Generation  \nMohammed Saim Ahmed Quadri∗, Yunzhe Xue†, Justin W. Ady‡, and Usman Roshan†  \n∗ Department of Computer Science, New Jersey Institute of Technology  \nNewark, NJ, USA  \n[mohammedsaimquadri@gmail.com](mohammedsaimquadri@gmail.com)  \n† Department of Data Science, New Jersey Institute of Technology  \nNewark, NJ, USA  \n[yunzhexue@gmail.com](yunzhexue@gmail.com); [usmanroshan@gmail.com](usmanroshan@gmail.com)  \n‡ Vascular and Endovascular Surgery, Robert Wood Johnson Hospital  \nNew Brunswick, NJ, USA  \n[jwa60@rwjms.rutgers.edu](jwa60@rwjms.rutgers.edu)  \nAbstract—Deploying Large Language Models (LLMs) in high-stakes clinical settings remains limited by structural hallucinations, weak deterministic reasoning over tabular patient data, and omissions in vector retrieval. This paper presents the architecture and validation of Medi-Gemma, a Clinical Decision Support System (CDSS) for wound pathology triage and workflow automation. The platform introduces a decoupled framework that separates clinical perception from data orchestration while preserving traceable reasoning. Medi-Gemma uses a multi-stage pipeline coordinated by a centralized ClinicalOrchestrator. Data requests are handled without generative inference by a DataManager that cleans unstructured Electronic Medical Record (EMR) files through type coercion. Natural language queries are processed by a hierarchical IntentRouter, which routes requests to deterministic analytics paths executed by a PandasQueryEngine or to patient-specific reasoning managed by a ClinicalRAGEngine using a CPU-optimized vector store. A key contribution is the Ground Truth Injection Module, which intercepts patient-specific queries, extracts numeric identification tokens, queries the structured dataframe via Pandas, retrievesthe latest validated clinical state, and embeds this snapshot as an overriding context block in the LLM prompt before generation. Safety compliance is enforced by a deterministic ProtocolManager that maps clinical terminology to fixed evidence-based risk pathways, while a SafetyVerifier phrase filter prevents output rule violations. Validation shows that this architecture eliminates semantic context drift, prevents database compilation crashes, and improves factual adherence to backend clinical repositories. These results support Medi-Gemma as a safer pattern for LLM-based clinical decision support where structured data fidelity, retrieval grounding, and deterministic safeguards are essential.  \nIndex Terms—Clinical Decision Support Systems, RetrievalAugmented Generation, Large Language Models, Electronic Medical Records, Clinical AI, Intent Routing, Ground Truth Injection, Agentic AI  \nI. INTRODUCTION  \nThe integration of Large Language Models (LLMs) into hospital workflows holds transformative potential for reducing clinician documentation overhead, generating structured  \nSubjective, Objective, Assessment, and Plan (SOAP) progress notes, and accelerating patient risk stratification [1], [2], [4],[5] . However, standard generative text architectures remain highly vulnerable to errors that limit their safe application in clinical environments. When processing unstructured narrative chart notes alongside structured longitudinal datasets (such as vital sign trends, lab results, and wound size dimensions), standard architectures frequently fail.  \nThese vulnerabilities stem from two primary architectural limitations that have also been identified in recent work on retrieval augmented medical language models and clinical AI systems [3], [5],[8]:  \n• Vector Retrieval Omissions: Standard RetrievalAugmented Generation (RAG) models split text files into arbitrary chunks and retrieve data based purely on semantic similarity scores. In longitudinal patient tracking, this can cause the system to overlook critical historical context or prioritize an outdated chart not","cbCait1IP2Hh0ive","https://ap.wps.com/l/cbCait1IP2Hh0ive","pdf",2814339,3,1,"English","en",105,"# Abstract\n# Introduction\n# Related Work","[{\"question\":\"Medi-Gemma如何缓解LLM在临床场景中的幻觉与错误推理？\",\"answer\":\"系统将生成式推理与确定性计算解耦：结构化数据用确定性分析引擎处理，患者最新验证状态通过Ground Truth Injection作为覆盖上下文注入LLM提示，从而减少语义漂移并提升基于记录的事实一致性。\"},{\"question\":\"Medi-Gemma的查询处理流程如何工作？\",\"answer\":\"数据由DataManager对非结构化EMR进行类型强制与清洗；自然语言请求由层级IntentRouter路由到确定性分析路径或由ClinicalRAGEngine使用CPU优化向量存储执行的患者特定推理路径。\"},{\"question\":\"系统如何确保安全合规并避免输出违反规则？\",\"answer\":\"ProtocolManager将临床术语映射到固定的循证风险路径；同时通过SafetyVerifier的短语过滤机制阻止输出规则违规，强化可控性与合规性。\"}]",1784191060,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"medi-gemma-hybrid-clinical-decision-support-system-integrating-deterministic-emr-analytics-and-retrieval-augmented-generation","",{"@graph":35,"@context":84},[36,52,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":20},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/medi-gemma-hybrid-clinical-decision-support-system-integrating-deterministic-emr-analytics-and-retrieval-augmented-generation/83863/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Medi-Gemma如何缓解LLM在临床场景中的幻觉与错误推理？","Question",{"text":74,"@type":75},"系统将生成式推理与确定性计算解耦：结构化数据用确定性分析引擎处理，患者最新验证状态通过Ground Truth Injection作为覆盖上下文注入LLM提示，从而减少语义漂移并提升基于记录的事实一致性。","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"Medi-Gemma的查询处理流程如何工作？",{"text":79,"@type":75},"数据由DataManager对非结构化EMR进行类型强制与清洗；自然语言请求由层级IntentRouter路由到确定性分析路径或由ClinicalRAGEngine使用CPU优化向量存储执行的患者特定推理路径。",{"name":81,"@type":72,"acceptedAnswer":82},"系统如何确保安全合规并避免输出违反规则？",{"text":83,"@type":75},"ProtocolManager将临床术语映射到固定的循证风险路径；同时通过SafetyVerifier的短语过滤机制阻止输出规则违规，强化可控性与合规性。","https://schema.org",{"og:url":50,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":50},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":21,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]