MEDRD: An AI-Driven Rare Disease Diagnosis Chatbot using Retrieval- Augmented Generation and Phenotype Reasoning

MEDRD: An AI-Driven Rare Disease Diagnosis Chatbot using Retrieval- Augmented Generation and Phenotype Reasoning | 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 Research Article MEDRD: An AI-Driven Rare Disease Diagnosis Chatbot using Retrieval- Augmented Generation and Phenotype Reasoning Madhumitha s, Shalini R, Pavithra S, Geetha Narasimhan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9344681/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Rare diseases are very hard to diagnose. They do not happen often and can show symptoms doctors may not be familiar with. Doctors do not always know much about them, which makes diagnosis difficult. Large computer programs called language models are good at understanding medical information and finding answers. However, they are not widely used in doctors’ systems because they can sometimes make up information and do not always give answers based on facts. To fix these problems, we created MEDRD, a computer program that helps doctors diagnose diseases. MEDRD uses language models and adds information from medical knowledge to give more accurate answers. It automatically finds symptoms, maps symptoms to medical terms using the Human Phenotype Ontology, and looks up information using special vectors. MEDRD works in steps. First, it finds diseases that could match the symptoms. Next, it compares the symptoms to diseases. Finally, it ranks diseases and gives a list of possible diagnoses. The system uses Fast API and works with GPT-4 and GPT-4o models. It also uses a database to quickly find relationships between symptoms and diseases. MEDRD can talk in more than ten languages and can be controlled with voice, making it easier for doctors to use. We tested MEDRD with rare disease cases covering more than 500 diseases. The results showed MEDRD was correct 79% of the time when it gave three possible diagnoses, which was 6.8% better than methods that only match symptoms. Clinical decision support conversational AI Human Phenotype Ontology (HPO) large language models (LLMs) medical reasoning multilingual medical AI phenotype-driven diagnosis rare diseases retrieval-augmented generation (RAG) vector-based knowledge retrieval voice-assisted healthcare systems Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 06 May, 2026 Reviews received at journal 27 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviewers agreed at journal 22 Apr, 2026 Reviewers invited by journal 21 Apr, 2026 Editor assigned by journal 15 Apr, 2026 Submission checks completed at journal 15 Apr, 2026 First submitted to journal 07 Apr, 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-9344681","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":630279471,"identity":"f38717bc-cda5-430c-8597-c2df798019a7","order_by":0,"name":"Madhumitha s","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIie2PsQrCMBRFEwSddC449BcUIVNtP8SlIRAXnV0c6hIXcdaPEARBHAsFp2DXQBzM4qyb3XwWJ6HFboI5kAsP7uG9IGSx/Co4glDwQuS9xllcReH5WEFBKMmzVHFXY2Oyw95t6QVxzCH1N/MEtky9QZHSUcNeryl1d32WxKFSs72koBz5OCpSHF5vY6HxVo2IokIzEoOCo6RQcVe8kWVCB2/lxEhqyhWkeB01haZvJfYhy5WOvNbaoLC14pMHFSwkCraEJX9x5xzf4bD+UrFdNxN+QNKhudymXvFhn9C8GX5bfxFUKVssFst/8AQpgGyOTWQLMwAAAABJRU5ErkJggg==","orcid":"","institution":"Kingston Engineering College","correspondingAuthor":true,"prefix":"","firstName":"Madhumitha","middleName":"","lastName":"s","suffix":""},{"id":630279472,"identity":"9c7d157a-3e44-43a1-9cfe-69813e13c164","order_by":1,"name":"Shalini R","email":"","orcid":"","institution":"Kingston Engineering College","correspondingAuthor":false,"prefix":"","firstName":"Shalini","middleName":"","lastName":"R","suffix":""},{"id":630279473,"identity":"57d3cf37-0f87-410a-ad39-71d8a2ebe5cd","order_by":2,"name":"Pavithra S","email":"","orcid":"","institution":"Kingston Engineering College","correspondingAuthor":false,"prefix":"","firstName":"Pavithra","middleName":"","lastName":"S","suffix":""},{"id":630279474,"identity":"a176cbdd-8305-437a-a112-376134821ccf","order_by":3,"name":"Geetha Narasimhan","email":"","orcid":"","institution":"Kingston Engineering College","correspondingAuthor":false,"prefix":"","firstName":"Geetha","middleName":"","lastName":"Narasimhan","suffix":""}],"badges":[],"createdAt":"2026-04-07 11:55:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9344681/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9344681/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108181765,"identity":"8c35e1ac-f3a4-4a51-a12e-5d8f46ba91ad","added_by":"auto","created_at":"2026-04-30 08:58:54","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":897282,"visible":true,"origin":"","legend":"","description":"","filename":"MEDRDManuscript4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9344681/v1_covered_e45a7002-ea73-437e-93d8-5805b95a0c6d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MEDRD: An AI-Driven Rare Disease Diagnosis Chatbot using Retrieval- Augmented Generation and Phenotype Reasoning","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-rare-diseases","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Journal of Rare Diseases](https://link.springer.com/journal/44162)","snPcode":"44162","submissionUrl":"https://submission.nature.com/new-submission/44162/3","title":"Journal of Rare Diseases","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Open","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Clinical decision support, conversational AI, Human Phenotype Ontology (HPO), large language models (LLMs), medical reasoning, multilingual medical AI, phenotype-driven diagnosis, rare diseases, retrieval-augmented generation (RAG), vector-based knowledge retrieval, voice-assisted healthcare systems","lastPublishedDoi":"10.21203/rs.3.rs-9344681/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9344681/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRare diseases are very hard to diagnose. They do not happen often and can show symptoms doctors may not be familiar with. Doctors do not always know much about them, which makes diagnosis difficult. Large computer programs called language models are good at understanding medical information and finding answers. However, they are not widely used in doctors\u0026rsquo; systems because they can sometimes make up information and do not always give answers based on facts. To fix these problems, we created MEDRD, a computer program that helps doctors diagnose diseases. MEDRD uses language models and adds information from medical knowledge to give more accurate answers. It automatically finds symptoms, maps symptoms to medical terms using the Human Phenotype Ontology, and looks up information using special vectors. MEDRD works in steps. First, it finds diseases that could match the symptoms. Next, it compares the symptoms to diseases. Finally, it ranks diseases and gives a list of possible diagnoses. The system uses Fast API and works with GPT-4 and GPT-4o models. It also uses a database to quickly find relationships between symptoms and diseases. MEDRD can talk in more than ten languages and can be controlled with voice, making it easier for doctors to use. We tested MEDRD with rare disease cases covering more than 500 diseases. The results showed MEDRD was correct 79% of the time when it gave three possible diagnoses, which was 6.8% better than methods that only match symptoms.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e","manuscriptTitle":"MEDRD: An AI-Driven Rare Disease Diagnosis Chatbot using Retrieval- Augmented Generation and Phenotype Reasoning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-29 07:07:55","doi":"10.21203/rs.3.rs-9344681/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-06T12:45:16+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-27T11:11:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"224486282724397110052229146119461331044","date":"2026-04-24T07:46:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"256499831694781485303118025927636078962","date":"2026-04-23T20:29:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"304090826196054994102722224429615421310","date":"2026-04-22T08:47:09+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-21T09:47:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-15T04:34:44+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-15T04:33:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Rare Diseases","date":"2026-04-07T11:41:21+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-rare-diseases","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Journal of Rare Diseases](https://link.springer.com/journal/44162)","snPcode":"44162","submissionUrl":"https://submission.nature.com/new-submission/44162/3","title":"Journal of Rare Diseases","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Open","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2516562d-10b9-4e9f-9a43-183fba642f54","owner":[],"postedDate":"April 29th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-06T12:45:16+00:00","index":22,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-29T07:07:55+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-29 07:07:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9344681","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9344681","identity":"rs-9344681","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}