[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-351912-105":59,"doc-detail-351912-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","detection-of-cancer-recurrence-from-thai-english-electronic-medical-records-using-sentence-embeddings","Detection of cancer recurrence from Thai-English electronic medical records using sentence embeddings","","Study developed and validated monolingual and bilingual SBERT transformer sentence-embedding models to detect cancer recurrence from Thai-English electronic medical records in Thai oncology hospitals. A multicentre dataset of 32,436 documents from 1,250 patients supported model development, with external validation using 9,244 documents from 384 patients across two hospitals. MetBERT and SBERT variants were benchmarked using precision-recall metrics, showing strong performance for distant versus no recurrence and robust external validation for bilingual SBERT.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/detection-of-cancer-recurrence-from-thai-english-electronic-medical-records-using-sentence-embeddings/351912/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/detection-of-cancer-recurrence-from-thai-english-electronic-medical-records-using-sentence-embeddings/351912.png","ImageObject",300,407,{"name":92,"@type":93},"Ben ","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-06","2026-09-22",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",13,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What is the main objective of the study?","Question",{"text":113,"@type":114},"To develop and validate monolingual and bilingual SBERT models for detecting cancer recurrence from Thai-English electronic medical records without translation overhead.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How were the datasets used for model development and validation?",{"text":118,"@type":114},"Model development used a multicentre dataset of 32,436 documents from 1,250 patients, and external validation used 9,244 documents from 384 patients across two Thai cancer hospitals.",{"name":120,"@type":111,"acceptedAnswer":121},"Which model performed best for different recurrence comparisons?",{"text":122,"@type":114},"MetBERT achieved the highest AUPRC for locoregional versus no recurrence and locoregional versus distant recurrence, monolingual-SBERT excelled for distant versus no recurrence, and bilingual-SBERT showed strong external validation for distant versus no recurrence.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},351912,1790155054,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":39,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":46,"preview_page_count":19,"preview_urls":145},2336478951081,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Open access Original research  \nDetection of cancer recurrence from Thai-English electronic medical records using sentence embeddings  \nEkapob Sangariyavanich  ,1,2 Wanchana Ponthongmak,2 Nawanan Theera-Ampornpunt,2 Nat Tangchitnob,2 Gareth J McKay,3 Ammarin Thakkinstian2  \nTo cite: Sangariyavanich E, Ponthongmak W, TheeraAmpornpunt N, et al. Detection of cancer recurrence from Thai-English electronic medical records using sentence embeddings.  \nBMJ Health Care Inform 2026;33:e101997 . doi:10 . 1136/ bmjhci-2025-101997  \n► Additional supplemental material is published online only. To view, please visit the journal online ([https://doi.org/10.1136/](https://doi.org/10.1136/)[ ](https://doi.org/10.1136/)[bmjhci-2025-101997](bmjhci-2025-101997)) .  \nReceived 18 December 2025 Accepted 21 June 2026  \n© Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group.  \n1National Cancer Institute of Thailand, Bangkok, Thailand 2Department of Clinical Epidemiology and Biostatistics, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand  \n3Centre for Public Health, Queen’s University Belfast, Belfast, UK  \nCorrespondence to  \nDr Wanchana Ponthongmak; [wanchana.pon@mahidol.ac.th](wanchana.pon@mahidol.ac.th)  \nABSTRACT  \nObjective This study developed and validated monolingual and bilingual sentence-bidirectional encoder representations from transformers (SBERT) models for detecting cancer recurrence within Thai-English electronic medical records (EMRs) from Thai cancer hospitals. Method A multicentre dataset of 32 436 documents from 1250 patients was used for model development. External validation involved an independent dataset of 9244 documents from 384 patients across two Thai cancer hospitals. Performance was benchmarked against a finetuned PubMedBERT (MetBERT) .  \nResults The development dataset included breast (43 .9%), colorectal (12 . 1%), cervical (28 .0%) and head and neck (16 .0%) cancers. MetBERT achieved the highest area under the precision-recall curve (AUPRC) for locoregional versus no recurrence (11 . 1%) and locoregional versus distant recurrence (91 .7%), while monolingual-SBERT excelled at distant versus no recurrence (32 .0%) .  \nExternal validation demonstrated MetBERT superiority for locoregional versus no recurrence (9 .30%–21 .50%) . For distant versus no recurrence, bilingual-SBERT performed best with AUPRC 17.55%–24 .39% . While MetBERT led in distinguishing locoregional versus distant recurrence (88 .30%–94 .70%), bilingual-SBERT demonstrated robust external validation performance (AUPRC 85 .25%–91 .80%) . Discussion Low AUPRC values (9%–32%) reflect the extreme class imbalance in real-world data (~1% recurrence prevalence) . Despite this, fine-tuned MetBERT achieved highest performance, while bilingual-SBERT demonstrated superior robustness during external validation. This validates sentence embedding models for handling mixed Thai-English medical records in multilingual clinical environments.  \nConclusion Sentence embedding frameworks provide a practical, generalisable solution for detecting cancer recurrence within multilingual EMRs. Despite textlength constraints, these models are suitable for clinical integration as a screening tool for cancer registry workflows.  \nINTRODUCTION  \nCancer continues to represent a significant global health burden, with 23.6 million new cases and 10 million deaths documented worldwide in 2019.1 Although therapeutic  \nWHAT IS ALREADY KNOWN ON THIS TOPIC  \n\n| ⇒ Manual recurrence detection is labour-intensive for registrars, and while BERT-based models show promise, they require more external validation in diverse clinical settings.\u003Cbr>WHAT THIS STUDY ADDS |\n| --- |\n| ⇒ This study validates sentence-bidirectional encoder representations from transformers (SBERT) models that process mixed Thai-English EMRs directly without translation overhead.\u003Cbr>⇒ It also provides a head-t","cbCairGkEy8PyJdm","https://ap.wps.com/l/cbCairGkEy8PyJdm","pdf",1920323,"English","# Abstract\n# Introduction\n# What is already known on this topic\n# What this study adds\n# How this study might affect research, practice or policy","[{\"question\":\"What is the main objective of the study?\",\"answer\":\"To develop and validate monolingual and bilingual SBERT models for detecting cancer recurrence from Thai-English electronic medical records without translation overhead.\"},{\"question\":\"How were the datasets used for model development and validation?\",\"answer\":\"Model development used a multicentre dataset of 32,436 documents from 1,250 patients, and external validation used 9,244 documents from 384 patients across two Thai cancer hospitals.\"},{\"question\":\"Which model performed best for different recurrence comparisons?\",\"answer\":\"MetBERT achieved the highest AUPRC for locoregional versus no recurrence and locoregional versus distant recurrence, monolingual-SBERT excelled for distant versus no recurrence, and bilingual-SBERT showed strong external validation for distant versus no recurrence.\"}]","Detection of cancer recurrence from Thai-English electronic medical records using sentence embeddings | PDF",1790096491,[146,150,154,158],{"page_no":8,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/60c49c615da7a08e3acd20ff83847f16.webp",983,1310,{"page_no":14,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/a50a6e4d9f85b64b16818775092271d7.webp",982,1309,{"page_no":81,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/4dcb3340f7bfa3f04bb7d57ab1a05c01.webp",981,1308,{"page_no":19,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/bb0d92fa6ba6148b4a0a69c225064e14.webp",980,1306]