Medical Image Captioning via Generative Pretrained Transformers

Medical Image Captioning via Generative Pretrained Transformers | 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 Medical Image Captioning via Generative Pretrained Transformers Alexander Selivanov, Oleg Rogov, Daniil Chesakov, Artem Shelmanov, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2197859/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 13 Mar, 2023 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract The automatic clinical caption generation problem is referred to as proposed model combining the analysis of frontal chest X-Rayscans with structured patient information from the radiology records. We combine two language models, the Show-Attend-Telland the GPT-3, to generate comprehensive and descriptive radiology records. The proposed combination of these models generates a textual summary with the essential information about pathologies found, their location, and the 2D heatmaps localizing each pathology on the original X-Ray scans. The proposed model is tested on two medical datasets, the Open-I,MIMIC-CXR, and the general-purpose MS-COCO. The results measured with the natural language assessment metrics provetheir efficient applicability to the chest X-Ray image captioning. Health sciences/Health care/Medical imaging/Radiography Physical sciences/Mathematics and computing/Computer science Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 13 Mar, 2023 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Major revision 03 Feb, 2023 Reviews received at journal 02 Feb, 2023 Reviews received at journal 10 Dec, 2022 Reviewers agreed at journal 30 Nov, 2022 Reviewers agreed at journal 29 Nov, 2022 Reviewers invited by journal 22 Nov, 2022 Editor assigned by journal 20 Nov, 2022 Editor invited by journal 07 Nov, 2022 Submission checks completed at journal 07 Nov, 2022 First submitted to journal 24 Oct, 2022 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-2197859","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":150129124,"identity":"a55a4f85-f17f-44c9-abc7-4d8a3db68068","order_by":0,"name":"Alexander Selivanov","email":"","orcid":"","institution":"Skolkovo Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Selivanov","suffix":""},{"id":150129125,"identity":"21d443bf-1883-45e2-b9ea-ebe6c976c71a","order_by":1,"name":"Oleg Rogov","email":"","orcid":"","institution":"Skolkovo Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Oleg","middleName":"","lastName":"Rogov","suffix":""},{"id":150129127,"identity":"aa9048fd-a7b7-434f-9665-b8ecdbe21c88","order_by":2,"name":"Daniil Chesakov","email":"","orcid":"","institution":"Skolkovo Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Daniil","middleName":"","lastName":"Chesakov","suffix":""},{"id":150129128,"identity":"101a0d09-aa60-47b5-937d-551fe183c6e2","order_by":3,"name":"Artem Shelmanov","email":"","orcid":"","institution":"Skolkovo Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Artem","middleName":"","lastName":"Shelmanov","suffix":""},{"id":150129130,"identity":"d85ee03a-0ebe-41cb-a776-08f3a90b3ef6","order_by":4,"name":"Irina Fedulova","email":"","orcid":"","institution":"Philips (Russia)","correspondingAuthor":false,"prefix":"","firstName":"Irina","middleName":"","lastName":"Fedulova","suffix":""},{"id":150129134,"identity":"9795154a-1d70-40b5-86be-6f96b344a5b0","order_by":5,"name":"Dmitry V. Dylov","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1ElEQVRIie3OMQrCMBTG8SeVTqGuKYK9QkUQh9KzpBTsYkUQpINDQehU6BG8Ql2cXwnUJd0de4SeQIw4uYS4Cea/vfD9IAAm0y+G4EMPMAMCI5Q30yNytpAEEL8hUa5NHLSulGU8OZcc+ZCFW5gKX0lctA+UCZ7WXcEQRbwHZ6MmPpIljYp1Wk+Ij02B8odEjyRe9SIPfRIw6EpJcg3icnu/YiKY16JlKNo4Koi9UxLndrrch4x6Xhk3fXYMo4pYtZKABWP68WCr9280aIxMJpPpn3sC11pMakzEyBgAAAAASUVORK5CYII=","orcid":"","institution":"Skolkovo Institute of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Dmitry","middleName":"V.","lastName":"Dylov","suffix":""}],"badges":[],"createdAt":"2022-10-24 08:59:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2197859/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2197859/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-023-31223-5","type":"published","date":"2023-03-13T20:02:33+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":28851042,"identity":"8ad2a6cc-9e38-4f30-936e-ee75ccbdcb3e","added_by":"auto","created_at":"2022-11-09 11:25:55","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7656221,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2197859/v1_covered.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Medical Image Captioning via Generative Pretrained Transformers","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"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-2197859/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2197859/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The automatic clinical caption generation problem is referred to as proposed model combining the analysis of frontal chest X-Rayscans with structured patient information from the radiology records. We combine two language models, the Show-Attend-Telland the GPT-3, to generate comprehensive and descriptive radiology records. The proposed combination of these models generates a textual summary with the essential information about pathologies found, their location, and the 2D heatmaps localizing each pathology on the original X-Ray scans. The proposed model is tested on two medical datasets, the Open-I,MIMIC-CXR, and the general-purpose MS-COCO. The results measured with the natural language assessment metrics provetheir efficient applicability to the chest X-Ray image captioning.","manuscriptTitle":"Medical Image Captioning via Generative Pretrained Transformers","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-11-09 11:25:38","doi":"10.21203/rs.3.rs-2197859/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2023-02-03T08:00:06+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2023-02-02T05:00:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2022-12-10T10:41:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"3475bc75-81dd-4fae-8068-cf315a13aa18","date":"2022-11-30T07:48:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"c14ffeba-1cae-4a2c-a803-c49771d489b0","date":"2022-11-29T13:20:20+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2022-11-23T01:54:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2022-11-20T10:19:02+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2022-11-07T13:26:26+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2022-11-07T13:20:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2022-10-24T08:53:37+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":"e57b4f37-62a2-432b-bc36-18ecb0f01722","owner":[],"postedDate":"November 9th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":16780798,"name":"Health sciences/Health care/Medical imaging/Radiography"},{"id":16780799,"name":"Physical sciences/Mathematics and computing/Computer science"}],"tags":[],"updatedAt":"2023-10-16T20:15:37+00:00","versionOfRecord":{"articleIdentity":"rs-2197859","link":"https://doi.org/10.1038/s41598-023-31223-5","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2023-03-13 20:02:33","publishedOnDateReadable":"March 13th, 2023"},"versionCreatedAt":"2022-11-09 11:25:38","video":"","vorDoi":"10.1038/s41598-023-31223-5","vorDoiUrl":"https://doi.org/10.1038/s41598-023-31223-5","workflowStages":[]},"version":"v1","identity":"rs-2197859","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2197859","identity":"rs-2197859","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}