Large Language Models in Portuguese for Healthcare: A Systematic Review

Large Language Models in Portuguese for Healthcare: A Systematic Review | 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 Systematic Review Large Language Models in Portuguese for Healthcare: A Systematic Review Andre Massahiro Shimaoka, Antonio Carlos da Silva Junior, José Marcio Duarte, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6673690/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 12 Mar, 2026 Read the published version in Research on Biomedical Engineering → Version 1 posted 10 You are reading this latest preprint version Abstract Purpose: This study addresses Large Language Models (LLMs) pre-trained in Portuguese for healthcare applications, focusing on contextual embeddings. Research on LLMs for natural language processing (NLP) tasks in Portuguese is limited, especially within healthcare. However, LLMs demonstrate potential in clinical decision support, diagnosis assistance, patient care, and other healthcare applications. In view thereof, the present work assesses the current state of LLMs in Portuguese for healthcare. Methods: Our Systematic Literature Review (SLR) followed standard protocols: search, screening based on inclusion/exclusion criteria, quality assessment, data extraction, and analysis. Results: We identified 28 models, mostly based on BERTimbau, mBERT, and BioBERTpt. Adaptation strategies such as fine-tuning, domain-adaptive pre-training, training from scratch, and zero-shot learning have been the most prevalent. Several datasets have been used, including clinical records, social media, and scientific repositories. LLMs in Portuguese are being applied in mental health, general medicine, COVID-19, oncology, and other related areas, accomplishing classification tasks, followed by named entity recognition (NER), topic modeling, question answering, text generation, and conversational agents. Conclusion: Our study identified gaps and opportunities: (1) base models such as LLAMA, T5, ELECTRA, BART, XLM-R, Falcon, Mistral, BLOOM are unexplored yet; (2) there is a lack of detailed fine-tuning specifications, hindering reproducibility; (3) many healthcare fields are not even tackled; (4) clinical and hospital data have been widely used but not shared; (5) social media data need caution because it can introduce inconsistencies; (6) data privacy, especially de-identification and anonymization, have been largely overlooked; and (7) Brazilian healthcare data present large opportunities. Large Language Models Natural Language Processing Healthcare Artificial Intelligence Portuguese Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 12 Mar, 2026 Read the published version in Research on Biomedical Engineering → Version 1 posted Editorial decision: Revision requested 05 Oct, 2025 Reviews received at journal 08 Sep, 2025 Reviews received at journal 22 Aug, 2025 Reviewers agreed at journal 19 Aug, 2025 Reviewers agreed at journal 19 Aug, 2025 Reviewers agreed at journal 19 Aug, 2025 Reviewers invited by journal 19 Aug, 2025 Editor assigned by journal 12 Jun, 2025 Submission checks completed at journal 16 May, 2025 First submitted to journal 15 May, 2025 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-6673690","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":457564067,"identity":"1a4da907-6be1-413a-8d78-2993849f0900","order_by":0,"name":"Andre Massahiro Shimaoka","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABD0lEQVRIiWNgGAWjYHACAxCRwAYkGBsYbBgYmBkfgEWI1ZIG1MJsQJwWBoiWw0CKgBb+2c3bPnxgqMvjEzv88OGMmvOJ29mZGR/+3MGQZ96AXYvEnWPFM2cwHC5mk04zNtxw7HbizmZmZmPeMwzFMgdwWHMjx5iZh+FAYpt0gpnkA7bbiRsO8x+TZmxjSJyBQ4c8SMsfhjqglvRvkg/+nQNqYWb/+ROPFgOQFqCPgVpyzCQ3th0AaWFj4MWjxfBGWjFjj8FhkJZiw5l9ycZALczSvGckiiVwaJG7kbyZ4UdFXeL82ekbH/Z8s5PdcP4w48efO2zycGmBOg9dgLEBvwYsABipo2AUjIJRMApgAAAHn1veV7Z20wAAAABJRU5ErkJggg==","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":true,"prefix":"","firstName":"Andre","middleName":"Massahiro","lastName":"Shimaoka","suffix":""},{"id":457564068,"identity":"392da553-e44c-4508-9498-02b163a31dc5","order_by":1,"name":"Antonio Carlos da Silva Junior","email":"","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Antonio","middleName":"Carlos da Silva","lastName":"Junior","suffix":""},{"id":457564069,"identity":"452684f5-9f14-4d78-ade0-32735a50972f","order_by":2,"name":"José Marcio Duarte","email":"","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"José","middleName":"Marcio","lastName":"Duarte","suffix":""},{"id":457564070,"identity":"a39d8e86-2d49-47e9-8ed3-cb655c983754","order_by":3,"name":"Thiago Bulhões da Silva Costa","email":"","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Thiago","middleName":"Bulhões da Silva","lastName":"Costa","suffix":""},{"id":457564071,"identity":"f21e88e1-d4fb-4965-adb8-da78100d4885","order_by":4,"name":"Ivan Torres Pisa","email":"","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Ivan","middleName":"Torres","lastName":"Pisa","suffix":""},{"id":457564072,"identity":"da07eb99-aa15-451f-a5e2-dc9ec668837a","order_by":5,"name":"Luciano Rodrigo Lopes","email":"","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Luciano","middleName":"Rodrigo","lastName":"Lopes","suffix":""},{"id":457564073,"identity":"86064924-a155-4f1b-9eb0-778d053eecf9","order_by":6,"name":"Paulo Bandiera-Paiva","email":"","orcid":"","institution":"Federal University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Paulo","middleName":"","lastName":"Bandiera-Paiva","suffix":""}],"badges":[],"createdAt":"2025-05-15 14:38:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6673690/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6673690/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s42600-026-00465-3","type":"published","date":"2026-03-12T16:00:12+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":104739862,"identity":"27f1733a-6152-4219-ac12-b2b544b24606","added_by":"auto","created_at":"2026-03-16 16:13:07","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":833506,"visible":true,"origin":"","legend":"","description":"","filename":"LLMSLR.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6673690/v1_covered_cc121dca-ab4f-41a1-a5a1-2bfc3c2fbe33.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Large Language Models in Portuguese for\nHealthcare: A Systematic Review","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"research-on-biomedical-engineering","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"rbme","sideBox":"Learn more about [Research on Biomedical Engineering](http://link.springer.com/journal/42600)","snPcode":"42600","submissionUrl":"https://submission.nature.com/new-submission/42600/3","title":"Research on Biomedical Engineering","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Large Language Models, Natural Language Processing, Healthcare, Artificial Intelligence, Portuguese","lastPublishedDoi":"10.21203/rs.3.rs-6673690/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6673690/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cb\u003ePurpose:\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis study addresses Large Language Models (LLMs) pre-trained in Portuguese for healthcare applications, focusing on contextual embeddings. Research on LLMs for natural language processing (NLP) tasks in Portuguese is limited, especially within healthcare. However, LLMs demonstrate potential in clinical decision support, diagnosis assistance, patient care, and other healthcare applications. In view thereof, the present work assesses the current state of LLMs in Portuguese for healthcare.\u003c/p\u003e\u003cp\u003e\u003cb\u003eMethods:\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOur Systematic Literature Review (SLR) followed standard protocols: search, screening based on inclusion/exclusion criteria, quality assessment, data extraction, and analysis.\u003c/p\u003e\u003cp\u003e\u003cb\u003eResults:\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe identified 28 models, mostly based on BERTimbau, mBERT, and BioBERTpt. Adaptation strategies such as fine-tuning, domain-adaptive pre-training, training from scratch, and zero-shot learning have been the most prevalent. Several datasets have been used, including clinical records, social media, and scientific repositories. LLMs in Portuguese are being applied in mental health, general medicine, COVID-19, oncology, and other related areas, accomplishing classification tasks, followed by named entity recognition (NER), topic modeling, question answering, text generation, and conversational agents.\u003c/p\u003e\u003cp\u003e\u003cb\u003eConclusion:\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOur study identified gaps and opportunities: (1) base models such as LLAMA, T5, ELECTRA, BART, XLM-R, Falcon, Mistral, BLOOM are unexplored yet; (2) there is a lack of detailed fine-tuning specifications, hindering reproducibility; (3) many healthcare fields are not even tackled; (4) clinical and hospital data have been widely used but not shared; (5) social media data need caution because it can introduce inconsistencies; (6) data privacy, especially de-identification and anonymization, have been largely overlooked; and (7) Brazilian healthcare data present large opportunities.\u003c/p\u003e","manuscriptTitle":"Large Language Models in Portuguese for\nHealthcare: A Systematic Review","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-22 03:59:18","doi":"10.21203/rs.3.rs-6673690/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-05T17:10:52+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-08T19:02:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-22T09:36:41+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"63732723498429476189090492471302006049","date":"2025-08-19T19:56:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"86464896824134590036972724639290427213","date":"2025-08-19T19:10:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"215732400753542853153202267617746087448","date":"2025-08-19T16:38:36+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-19T13:41:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-12T19:53:16+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-16T10:22:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Research on Biomedical Engineering","date":"2025-05-15T14:34:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"research-on-biomedical-engineering","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"rbme","sideBox":"Learn more about [Research on Biomedical Engineering](http://link.springer.com/journal/42600)","snPcode":"42600","submissionUrl":"https://submission.nature.com/new-submission/42600/3","title":"Research on Biomedical Engineering","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"575be2c1-56b1-4855-b0e2-b8df370af2db","owner":[],"postedDate":"May 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-16T16:06:13+00:00","versionOfRecord":{"articleIdentity":"rs-6673690","link":"https://doi.org/10.1007/s42600-026-00465-3","journal":{"identity":"research-on-biomedical-engineering","isVorOnly":false,"title":"Research on Biomedical Engineering"},"publishedOn":"2026-03-12 16:00:12","publishedOnDateReadable":"March 12th, 2026"},"versionCreatedAt":"2025-05-22 03:59:18","video":"","vorDoi":"10.1007/s42600-026-00465-3","vorDoiUrl":"https://doi.org/10.1007/s42600-026-00465-3","workflowStages":[]},"version":"v1","identity":"rs-6673690","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6673690","identity":"rs-6673690","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}