ADMEDTAGGER: an annotation framework for distillation of expert knowledge for the Polish medical language

ADMEDTAGGER: an annotation framework for distillation of expert knowledge for the Polish medical language | 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 Article ADMEDTAGGER: an annotation framework for distillation of expert knowledge for the Polish medical language Franciszek Górski, Andrzej Czyżewski This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8520529/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract In this work, we present an annotation framework that demonstrates how a multilingual LLM pretrained on a large corpus can be used as a teacher model to distill the expert knowledge needed for tagging medical texts in Polish. This work is part of a larger project called ADMEDVOICE, within which we collected an extensive corpus of medical texts representing five clinical categories - Radiology, Oncology, Cardiology, Hypertension, and Pathology. Using this data, we had to develop a multi-class classifier, but the fundamental problem turned out to be the lack of resources for annotating an adequate number of texts. Therefore, in our solution, we used the multilingual Llama3.1 model to annotate an extensive corpus of medical texts in Polish. Using our limited annotation resources, we verified only a portion of these labels, creating a test set from them. The data annotated in this way were then used for training and validation of 3 different types of classifiers based on the BERT architecture - the distilled DistilBERT model, BioBERT fine-tuned on medical data, and HerBERT fine-tuned on the Polish language corpus. Among the models we trained, the DistilBERT model achieved the best results, reaching an F1 score > 0.80 for each clinical category and an F1 score > 0.93 for 3 of them. In this way, we obtained a series of highly effective classifiers that represent an alternative to large language models, due to their nearly 500 times smaller size, 300 times lower GPU VRAM consumption, and several hundred times faster inference. Biological sciences/Computational biology and bioinformatics Health sciences/Health care Physical sciences/Mathematics and computing Health sciences/Medical research Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 10 Mar, 2026 Reviews received at journal 09 Mar, 2026 Reviewers agreed at journal 01 Mar, 2026 Reviews received at journal 16 Feb, 2026 Reviewers agreed at journal 16 Feb, 2026 Reviewers agreed at journal 02 Feb, 2026 Reviewers invited by journal 30 Jan, 2026 Editor assigned by journal 22 Jan, 2026 Editor invited by journal 20 Jan, 2026 Submission checks completed at journal 17 Jan, 2026 First submitted to journal 17 Jan, 2026 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-8520529","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":586985517,"identity":"d4365497-35ae-4a4c-9223-61971b49b8ec","order_by":0,"name":"Franciszek Górski","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABH0lEQVRIiWNgGAWjYBAC9gYQaQDEPBABAzaIEDMIN2DTwnMAQwtECKSFEY8WBiQtDBIJBLRIH3724EdBnT0/zwHmFx/31BnzSb49/IKhwjqxQboRuxa+NHPDHoPDzJK9DWyWM54dNmOTzkuzYDiTntggcxCrFnseBjNpBoMDbAbnGdiMeQ4csGGTzjEzYGw7nNggkYjdFh72b0AtdTz2EC11NmySZ4Ba/uHTwgOyhVnCgLeB+THQP2ZsEjzGDxgb8GopkwT6xUDizME2xhkHDhuz8eSlMSQcSzduw+EXoMO2Sfz4AwyxnuTDHz4cqDOc334WyKixlu2Xbj6ATQsSYGyTgBrDBo4aNgkCGoCA+QNUC5RBhJZRMApGwSgYEQAAAtBYsOyvtL4AAAAASUVORK5CYII=","orcid":"","institution":"Gdańsk University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Franciszek","middleName":"","lastName":"Górski","suffix":""},{"id":586985518,"identity":"c605402d-5fbc-4ab8-b34b-8433feb51021","order_by":1,"name":"Andrzej Czyżewski","email":"","orcid":"","institution":"Gdańsk University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Andrzej","middleName":"","lastName":"Czyżewski","suffix":""}],"badges":[],"createdAt":"2026-01-05 11:08:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8520529/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8520529/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102296769,"identity":"3154e3b0-16dc-46c8-830e-933b6c0c3fa6","added_by":"auto","created_at":"2026-02-10 10:21:32","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":408310,"visible":true,"origin":"","legend":"","description":"","filename":"ADMEDTAGGERSCIENTIFICrevisedv2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8520529/v1_covered_7f772880-be20-4238-b181-c6c929a9a911.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"ADMEDTAGGER: an annotation framework for distillation of expert knowledge for the Polish medical language","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8520529/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8520529/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"In this work, we present an annotation framework that demonstrates how a multilingual LLM pretrained on a large corpus can be used as a teacher model to distill the expert knowledge needed for tagging medical texts in Polish. This work is part of a larger project called ADMEDVOICE, within which we collected an extensive corpus of medical texts representing five clinical categories - Radiology, Oncology, Cardiology, Hypertension, and Pathology. Using this data, we had to develop a multi-class classifier, but the fundamental problem turned out to be the lack of resources for annotating an adequate number of texts. Therefore, in our solution, we used the multilingual Llama3.1 model to annotate an extensive corpus of medical texts in Polish. Using our limited annotation resources, we verified only a portion of these labels, creating a test set from them. The data annotated in this way were then used for training and validation of 3 different types of classifiers based on the BERT architecture - the distilled DistilBERT model, BioBERT fine-tuned on medical data, and HerBERT fine-tuned on the Polish language corpus. Among the models we trained, the DistilBERT model achieved the best results, reaching an F1 score \u003e 0.80 for each clinical category and an F1 score \u003e 0.93 for 3 of them. In this way, we obtained a series of highly effective classifiers that represent an alternative to large language models, due to their nearly 500 times smaller size, 300 times lower GPU VRAM consumption, and several hundred times faster inference.","manuscriptTitle":"ADMEDTAGGER: an annotation framework for distillation of expert knowledge for the Polish medical language","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-09 05:45:56","doi":"10.21203/rs.3.rs-8520529/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-10T05:57:30+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-09T15:27:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"293752721829250351773665252333094863983","date":"2026-03-02T04:08:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-16T22:47:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"43179720924568610505334522146820030039","date":"2026-02-16T21:21:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"271930637978084572046608130146700587142","date":"2026-02-02T15:35:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-30T08:30:17+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-22T08:02:14+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-20T14:48:34+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-17T14:32:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-17T14:30:02+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8e1098a7-77d7-4dfb-8e7a-d1b558adfa69","owner":[],"postedDate":"February 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62455758,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":62455759,"name":"Health sciences/Health care"},{"id":62455760,"name":"Physical sciences/Mathematics and computing"},{"id":62455761,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-18T06:40:06+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-09 05:45:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8520529","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8520529","identity":"rs-8520529","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}