Toward Interpretable Glucose Forecasting for Type 2 Diabetes: A Comparative Study among Traditional, Deep, and Large Language Models

Toward Interpretable Glucose Forecasting for Type 2 Diabetes: A Comparative Study among Traditional, Deep, and Large Language Models | 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 Toward Interpretable Glucose Forecasting for Type 2 Diabetes: A Comparative Study among Traditional, Deep, and Large Language Models Rawan Alredaini, Maysoon Abulkhair, Hind Almisbahi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7698906/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted 13 You are reading this latest preprint version Abstract Type 2 diabetes mellitus (T2DM) is a prevalent chronic condition characterized by elevated blood glucose levels resulting from insulin resistance or inadequate insulin secretion. Accurate prediction of future glucose levels is essential for minimizing complications and enabling proactive management. While machine learning and deep learning models have been extensively applied in this domain, the potential of large language models (LLMs) remains underexplored, with no prior studies systematically comparing them to conventional approaches using real patient data. In this study, we evaluate three model types: traditional (XGBoost, Random Forest), deep learning (GRU, LSTM, Transformer, Ensemble), and finetuned LLMs (GPT-4.1, MiniGPT, LLaMA-1B, LLaMA-7B), for predicting glucose levels 30, 60, and 90 minutes ahead using hybrid inputs of six static features and 20 prior CGM readings. GPT-4.1 achieved the best performance at 30 and 60 minutes, while LLaMA-7B excelled at 90 minutes. Among conventional models, LSTM showed the best performance. Beyond forecasting, interpretability was a central focus. We used explainable AI (XAI) techniques to interpret LSTM results, while GPT-4.1 explained its predictions directly using natural language, without additional training. Notably, the study revealed an alignment between the two explanation techniques, with both highlighting recent glucose readings as key predictors across all forecasting horizons. Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Health sciences/Endocrinology Full Text Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.pdf Cite Share Download PDF Status: Published Journal Publication published 16 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 06 Oct, 2025 Reviews received at journal 05 Oct, 2025 Reviewers agreed at journal 05 Oct, 2025 Reviewers agreed at journal 03 Oct, 2025 Reviewers agreed at journal 03 Oct, 2025 Reviews received at journal 03 Oct, 2025 Reviewers agreed at journal 03 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers invited by journal 02 Oct, 2025 Editor invited by journal 29 Sep, 2025 Editor assigned by journal 27 Sep, 2025 Submission checks completed at journal 26 Sep, 2025 First submitted to journal 23 Sep, 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-7698906","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":525567155,"identity":"a2d63a86-4a60-4dc6-883a-d448b38c78cd","order_by":0,"name":"Rawan Alredaini","email":"data:image/png;base64,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","orcid":"","institution":"King Abdulaziz University","correspondingAuthor":true,"prefix":"","firstName":"Rawan","middleName":"","lastName":"Alredaini","suffix":""},{"id":525567156,"identity":"25bd77f6-949d-4f11-ac10-d80a7325fcf6","order_by":1,"name":"Maysoon Abulkhair","email":"","orcid":"","institution":"King Abdulaziz University","correspondingAuthor":false,"prefix":"","firstName":"Maysoon","middleName":"","lastName":"Abulkhair","suffix":""},{"id":525567157,"identity":"164e2af9-289c-47cc-aea1-78c6490235fa","order_by":2,"name":"Hind Almisbahi","email":"","orcid":"","institution":"King Abdulaziz University","correspondingAuthor":false,"prefix":"","firstName":"Hind","middleName":"","lastName":"Almisbahi","suffix":""}],"badges":[],"createdAt":"2025-09-24 03:23:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7698906/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7698906/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-32373-4","type":"published","date":"2025-12-16T15:58:33+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93609275,"identity":"cabb5734-acfc-4fa6-96fa-1e2316881397","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5753,"visible":true,"origin":"","legend":"","description":"","filename":"801cb118efaf4113bba69e66b4b92398.json","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/b14030025547e0c9f6eda7a7.json"},{"id":93609276,"identity":"1719d163-7443-4fe0-bc9a-72325d22717e","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94911,"visible":true,"origin":"","legend":"","description":"","filename":"CGMdatapreprocessingwindow.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/848a257ce41a6bcfdfd5930f.png"},{"id":93609639,"identity":"4653ce31-8337-44f9-a246-b9febc500abd","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78322,"visible":true,"origin":"","legend":"","description":"","filename":"Methodologydiagram.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/f9bc635161fcfd7338c566c3.png"},{"id":93609641,"identity":"ce0483ed-849c-4935-ac7b-48b3d5e1b8b5","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97672,"visible":true,"origin":"","legend":"","description":"","filename":"SHAP30min.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/474e5b735960b0e89cb2c7b9.png"},{"id":93609278,"identity":"14df445d-12cb-4ee4-8dc2-39c9fa4fb65c","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98421,"visible":true,"origin":"","legend":"","description":"","filename":"SHAP60min.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/fa618caada8b72af3274c4e4.png"},{"id":93609279,"identity":"8b46a6e2-0872-4a30-b687-74efab816d3d","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":100211,"visible":true,"origin":"","legend":"","description":"","filename":"SHAP90min.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/ae0d0cc427faff88676bf890.png"},{"id":93609280,"identity":"9dc6da16-bc21-42e1-a4d6-c3106224e449","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":74937,"visible":true,"origin":"","legend":"","description":"","filename":"SHAPWaterfall.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/a588167948bae957ac232cef.png"},{"id":93609303,"identity":"bb94725d-21e8-4f65-93a2-b95356de870f","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7102586,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/8c1071cf619e80d3a8c0318c.pdf"},{"id":93609640,"identity":"3a54bc4b-7984-4bb9-8656-d91373217a87","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49037,"visible":true,"origin":"","legend":"","description":"","filename":"coverletterScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/114cc18ac0807f8b75f11368.pdf"},{"id":93609282,"identity":"122c2a4b-ccd1-464a-923d-317ac01672ba","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":198022,"visible":true,"origin":"","legend":"","description":"","filename":"datapreprocessingdiagram.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/f54ecde44a45ead6d2c3b8d1.png"},{"id":93611861,"identity":"b55d1b7f-2273-4b9a-8086-3e94ecbe29a2","added_by":"auto","created_at":"2025-10-15 16:15:55","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":296836,"visible":true,"origin":"","legend":"","description":"","filename":"gpt4resultpic.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/5a5c35c0fb18770841a1dc76.png"},{"id":93609289,"identity":"e30d1970-a3a6-4971-a958-3b019ab1f67f","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"ldf","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49107,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaall.ldf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/b94c2355b4381ea758eca8b7.ldf"},{"id":93609283,"identity":"8a7d230a-d2fb-4a7c-913a-1971500f1799","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"ldf","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268657,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaen.ldf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/029e94d2b525a7e1492d6569.ldf"},{"id":93611860,"identity":"73950f3e-8c98-493b-a4d5-23cabf9311f9","added_by":"auto","created_at":"2025-10-15 16:15:55","extension":"sty","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":15480,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrv.sty","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/c743e273c43e7a27e3fb07a1.sty"},{"id":93609645,"identity":"8272b200-3548-4c48-a67b-ea12fad11720","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"bbl","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16032,"visible":true,"origin":"","legend":"","description":"","filename":"main.bbl","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/a9ae5c29f0897de2c2f27950.bbl"},{"id":93609298,"identity":"36e466d5-0fe9-4305-9657-b7f0a10a2de3","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1246256,"visible":true,"origin":"","legend":"","description":"","filename":"myoverleafprojecttoScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/c4050b2e68fb2679e2b6e50f.pdf"},{"id":93611862,"identity":"ae180b8f-dea7-4b68-8bbb-40b266d194af","added_by":"auto","created_at":"2025-10-15 16:15:55","extension":"bst","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36153,"visible":true,"origin":"","legend":"","description":"","filename":"naturemagdoi.bst","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/ad486d59fcc73bbbacc7356e.bst"},{"id":93609291,"identity":"b9ce6af4-72bb-4a62-ab32-88c19964bee6","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":88968,"visible":true,"origin":"","legend":"","description":"","filename":"output30min1.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/fd508b7af15e54e7ee966101.png"},{"id":93609643,"identity":"375c7b4c-3b94-45d6-99bc-d46586b8c13b","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18823,"visible":true,"origin":"","legend":"","description":"","filename":"output30min2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/f17f796df56aabd4c1782478.png"},{"id":93609295,"identity":"92920ca9-3c53-4bac-8fad-82dddc4d858a","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90610,"visible":true,"origin":"","legend":"","description":"","filename":"output60min1.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/76274a03266d6ded6c6ec023.png"},{"id":93611864,"identity":"2074af60-519f-4d08-b91b-16ad4f5cf370","added_by":"auto","created_at":"2025-10-15 16:15:55","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17935,"visible":true,"origin":"","legend":"","description":"","filename":"output60min2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/b9f0115eaf574f42d0081c4c.png"},{"id":93609651,"identity":"af79ca30-2a4f-47d8-8b08-cff583e58da8","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":92575,"visible":true,"origin":"","legend":"","description":"","filename":"output90min1.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/9f779ad0115ab9748e116997.png"},{"id":93609286,"identity":"44e8bf24-4c2c-4f15-909e-d37d8905ac18","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17973,"visible":true,"origin":"","legend":"","description":"","filename":"output90min2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/9fdbcad5b4ef5b5abaf84521.png"},{"id":93609300,"identity":"25dc45fc-f1ce-45a5-978a-a52423b7248a","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"jpg","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":379752,"visible":true,"origin":"","legend":"","description":"","filename":"stream.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/4eedd759280eab3defa9d310.jpg"},{"id":93609292,"identity":"c91ab4f2-550b-4761-81f6-eb7eb22348ef","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108831,"visible":true,"origin":"","legend":"","description":"","filename":"time2vecLSTMdiagram.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/8cab90be4e6297e0db052f00.png"},{"id":93609646,"identity":"d77ff74d-3324-4cb0-b94c-112629b52253","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119804,"visible":true,"origin":"","legend":"","description":"","filename":"transformerdiagram.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/0da6dca6829e03706eb5caf9.png"},{"id":93609306,"identity":"9a4840a5-d9db-4eb5-a1e8-c742ac657f02","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"cls","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5824,"visible":true,"origin":"","legend":"","description":"","filename":"wlscirep.cls","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/bb13811292217a4553630d03.cls"},{"id":93609653,"identity":"d35e02ea-d47a-4249-8cff-b593bf9e2762","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18823,"visible":true,"origin":"","legend":"","description":"","filename":"output30min2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/261ffb7274589e06f3b9dfeb.png"},{"id":93609648,"identity":"dba639b8-8371-4db6-acde-ef9542f04c7c","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18019,"visible":true,"origin":"","legend":"","description":"","filename":"xx60min2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/571f059ebf6e9fb9ae3cc32c.png"},{"id":93609650,"identity":"6273577e-ba7b-4978-b721-bdd165337998","added_by":"auto","created_at":"2025-10-15 16:07:55","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18123,"visible":true,"origin":"","legend":"","description":"","filename":"xx90min2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/f73eb8d4aaf443b2b6e83bfd.png"},{"id":93609304,"identity":"c8f5b235-60de-41ff-a4c7-84d0321730c6","added_by":"auto","created_at":"2025-10-15 15:59:55","extension":"xml","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105789,"visible":true,"origin":"","legend":"","description":"","filename":"801cb118efaf4113bba69e66b4b923981structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/ba414dde0f2c0b6eb9e9860e.xml"},{"id":98815357,"identity":"b8d277fb-33b9-479d-9d96-cc2fc6848703","added_by":"auto","created_at":"2025-12-22 16:14:26","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1181083,"visible":true,"origin":"","legend":"","description":"","filename":"myoverleafprojecttoScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1_covered_ddfc030e-4d51-4a77-8c17-ca9c3673f597.pdf"},{"id":93612787,"identity":"4b8713bf-b97a-496b-b9ba-f49864621f7d","added_by":"auto","created_at":"2025-10-15 16:24:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":7221826,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698906/v1/2640e25d097d3b86a5a82480.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eToward Interpretable Glucose Forecasting for Type 2 Diabetes: A Comparative Study among Traditional, Deep, and Large Language Models\u003c/p\u003e","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-7698906/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7698906/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Type 2 diabetes mellitus (T2DM) is a prevalent chronic condition characterized by elevated blood glucose levels resulting from insulin resistance or inadequate insulin secretion. Accurate prediction of future glucose levels is essential for minimizing complications and enabling proactive management. While machine learning and deep learning models have been extensively applied in this domain, the potential of large language models (LLMs) remains underexplored, with no prior studies systematically comparing them to conventional approaches using real patient data. In this study, we evaluate three model types: traditional (XGBoost, Random Forest), deep learning (GRU, LSTM, Transformer, Ensemble), and finetuned LLMs (GPT-4.1, MiniGPT, LLaMA-1B, LLaMA-7B), for predicting glucose levels 30, 60, and 90 minutes ahead using hybrid inputs of six static features and 20 prior CGM readings. GPT-4.1 achieved the best performance at 30 and 60 minutes, while LLaMA-7B excelled at 90 minutes. Among conventional models, LSTM showed the best performance. Beyond forecasting, interpretability was a central focus. We used explainable AI (XAI) techniques to interpret LSTM results, while GPT-4.1 explained its predictions directly using natural language, without additional training. Notably, the study revealed an alignment between the two explanation techniques, with both highlighting recent glucose readings as key predictors across all forecasting horizons.","manuscriptTitle":"Toward Interpretable Glucose Forecasting for Type 2 Diabetes: A Comparative Study among Traditional, Deep, and Large Language Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-15 15:59:50","doi":"10.21203/rs.3.rs-7698906/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-06T18:30:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-05T14:43:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"13193909042307505790369750692267047321","date":"2025-10-05T13:47:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"235872923607970345301890106489305769307","date":"2025-10-03T11:16:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"243148347893032074868697222312494284928","date":"2025-10-03T07:09:10+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-03T06:23:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"90871317111593303247331077546699299059","date":"2025-10-03T05:57:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"54182781644588462301120219764417989648","date":"2025-10-03T03:41:03+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-02T19:43:43+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-29T19:22:56+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-27T04:29:09+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-26T09:15:15+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-09-24T03:19:25+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":"00e43a53-b70f-4092-bc23-cf729abf8ba7","owner":[],"postedDate":"October 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":56359536,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":56359537,"name":"Health sciences/Diseases"},{"id":56359538,"name":"Health sciences/Endocrinology"}],"tags":[],"updatedAt":"2025-12-22T16:11:45+00:00","versionOfRecord":{"articleIdentity":"rs-7698906","link":"https://doi.org/10.1038/s41598-025-32373-4","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-12-16 15:58:33","publishedOnDateReadable":"December 16th, 2025"},"versionCreatedAt":"2025-10-15 15:59:50","video":"","vorDoi":"10.1038/s41598-025-32373-4","vorDoiUrl":"https://doi.org/10.1038/s41598-025-32373-4","workflowStages":[]},"version":"v1","identity":"rs-7698906","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7698906","identity":"rs-7698906","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}