[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-426313-105":3,"detail-sidebar-cat-0-en-105":80,"doc-detail-426313-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":73,"head_meta":75,"extra_data":77,"updated_unix":79},105,"en","hybrid-value-aware-transformer-identifies-novel-and-suspected-drugs-for-alzheimers-disease","HYBRID VALUE-AWARE TRANSFORMER IDENTIFIES NOVEL AND SUSPECTED DRUGS FOR ALZHEIMER’S DISEASE","","Drug discovery for Alzheimer’s disease and related dementias (ADRD) is hindered by the complex, multifactorial nature of these conditions. A research team introduced the Hybrid Value-Aware Transformer (HVAT), a deep learning architecture for multimodal health data that extends Transformer models with clinical tokens and incorporates numerical values from labs and cumulative drug doses. Using a large U.S. Veterans case-control cohort, HVAT distinguished ADRD cases from controls with AUC=0.775, and produced explainable population-level drug impact scores that highlight dose-dependent associations and enable detection of both suspected and novel drug candidates.",{"@graph":14,"@context":72},[15,34,55],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/hybrid-value-aware-transformer-identifies-novel-and-suspected-drugs-for-alzheimers-disease/426313/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/hybrid-value-aware-transformer-identifies-novel-and-suspected-drugs-for-alzheimers-disease/426313.png","ImageObject",300,407,{"name":42,"@type":43},"Grenda","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-16","2026-09-29",true,{"@type":52,"interactionType":53,"userInteractionCount":33},"InteractionCounter",{"@type":54},"ViewAction",{"@type":56,"mainEntity":57},"FAQPage",[58,64,68],{"name":59,"@type":60,"acceptedAnswer":61},"What is the Hybrid Value-Aware Transformer (HVAT) designed to do?","Question",{"text":62,"@type":63},"HVAT is a deep learning architecture that models multimodal health data to identify dose-dependent associations between drug exposures and ADRD risk.","Answer",{"name":65,"@type":60,"acceptedAnswer":66},"How was HVAT evaluated in the study?",{"text":67,"@type":63},"HVAT was trained on a case-control cohort of roughly one million U.S. Veterans aged 65+ with up to 10 years of medical history, achieving AUC=0.775 in distinguishing ADRD cases from controls.",{"name":69,"@type":60,"acceptedAnswer":70},"How does the study make the findings interpretable?",{"text":71,"@type":63},"It derives population-level drug impact scores using an explainable AI measure comparable to an odds ratio to quantify associations between normalized cumulative dose and ADRD risk.","https://schema.org",{"og:url":32,"og:type":74,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":76,"canonical":32},"index,follow",{"doc_id":78,"site_id":7},426313,1791313916,{"code":4,"msg":81,"data":82},"success",[83,87,91,95,100,105,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":84,"show_sort_weight":85,"slug":86},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":88,"show_sort_weight":89,"slug":90},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":92,"show_sort_weight":93,"slug":94},"Exam",70,"exam",{"id":96,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},5,"Comic",60,"comic",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},6,"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},7,"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},8,30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":96,"slug":129},19,"General","general",{"code":4,"msg":81,"data":131},{"doc_id":78,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":111,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":33,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":22,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":143,"read_time":30},7971474921005,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","1182 Innovation in Aging, 2025, Vol. 9, No. S2  \nAbstract citation ID: igaf122.3958  \nHYBRID VALUE-AWARE TRANSFORMER IDENTIFIES  \nNOVEL AND SUSPECTED DRUGS FOR  \nALZHEIMER’S DISEASE  \nQing Zeng1, Yijun Shao2, Ying Yin3, Edward Zamrini4, Ali Ahmed5, and Cheng Yan6, 1. The George Washington University, Washington, District of Columbia, United States, 2. George Washington University, Washington, DC, Washington, District of Columbia, United States, 3. George Washington University, George Washington University, District of Columbia, United States, 4. Irvine Clinical Research, Irvine, California, United States, 5. Washington DC VA Medical Center, Washington, District of Columbia, United States, 6. George Washington University, Washington, District of Columbia, United States  \nDrug discovery for Alzheimer’s disease and related dementias (ADRD) faces many challenges including the complex and multifactorial nature of ADRD. Our research team developed the Hybrid Value-Aware Transformer (HVAT), a novel deep learning architecture designed to model multimodal health data. HVAT extends Transformer models, a powerful AI technology originally developed for natural language processing and generation, with“clinical tokens” that represent both longitudinal and non-longitudinal data, while also incorporating numerical values such as laboratory results and cumulative drug doses. We used this hybrid, value-aware model to identify dose-dependent associations between drug exposures and ADRD risk. Specifically, we trained an HVAT model on a case-control cohort of approximately one million U.S. Veterans aged 65 years and older, with up to 10 years of medical history. HVAT achieved fair predictive performance (area under the curve [AUC] = 0.775) in distinguishing ADRD cases from controls. We then derived population-level drug impact scores, a novel explainable AI measure similar to odds ratio, to quantify the association between normalized cumulative dose andADRD risk. Our analyses yielded both suspected candidates, such as varenicline, which has supportive preclinical and early clinical evidence in ADRD, and novel candidates, such as metolazone, a thiazide-like diuretic not previously linked to dementia. These findings demonstrate that HVAT can recover known signals while also uncovering new drug repurposing opportunities. By combining deep learning with explainable AI analysis, this project represents a generalizable AI framework for accelerating drug discovery in aging research.","cbCailY4IW0gBJOY","https://ap.wps.com/l/cbCailY4IW0gBJOY","pdf",1040106,"English","# Abstract\n## Model design and data representation\n## Cohort training and predictive performance\n## Explainable drug impact scoring\n## Candidate recovery and novelty","[{\"question\":\"What is the Hybrid Value-Aware Transformer (HVAT) designed to do?\",\"answer\":\"HVAT is a deep learning architecture that models multimodal health data to identify dose-dependent associations between drug exposures and ADRD risk.\"},{\"question\":\"How was HVAT evaluated in the study?\",\"answer\":\"HVAT was trained on a case-control cohort of roughly one million U.S. Veterans aged 65+ with up to 10 years of medical history, achieving AUC=0.775 in distinguishing ADRD cases from controls.\"},{\"question\":\"How does the study make the findings interpretable?\",\"answer\":\"It derives population-level drug impact scores using an explainable AI measure comparable to an odds ratio to quantify associations between normalized cumulative dose and ADRD risk.\"}]","HYBRID VALUE-AWARE TRANSFORMER IDENTIFIES NOVEL AND SUSPECTED DRUGS FOR ALZHEIMER’S DISEASE | PDF",1790640166]