Comparison of AI-Generated Radiology Impressions: A Multi-Stakeholder Evaluation

Comparison of AI-Generated Radiology Impressions: A Multi-Stakeholder Evaluation | 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 Comparison of AI-Generated Radiology Impressions: A Multi-Stakeholder Evaluation Sharang Phadke, Nivedita Suresh, Zachary Allen, Anjali Balagopal, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8476600/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 Apr, 2026 Read the published version in npj Digital Medicine → Version 1 posted 11 You are reading this latest preprint version Abstract Objective To evaluate the quality, safety, and clinical utility of AI-generated radiology impressions compared with human-authored impressions across multiple clinical stakeholder groups. Materials & Methods A retrospective, blinded evaluation was conducted using 200 oncologic computed-tomography reports from a U.S. academic cancer center. Three impression types were assessed for each report: original radiologist-authored impressions, impressions generated by a custom domain-specific AI model fine-tuned on institutional data, and impressions generated by a general-purpose large language model. Original authoring radiologists, independent radiologists, and oncologists evaluated impressions using structured Likert-scale metrics assessing completeness, correctness, conciseness, clarity, clinical utility, and potential patient harm. Pairwise comparisons were performed using Wilcoxon signed-rank and two-proportion z-tests. Results Custom model AI impressions demonstrated near parity with human-authored impressions across most quality metrics. Original radiologists rated their own impressions as slightly more complete, while independent radiologists showed no significant differences between original and custom model impressions. Generic model impressions were longer, rated as more complete but significantly less concise. Patient harm ratings were uniformly low. Radiologists preferred original and custom model impressions over generic model impressions, whereas oncologists showed no significant preference. Discussion Evaluation outcomes varied by stakeholder group, highlighting differing priorities between radiologists and oncologists. Low inter-rater agreement across several quality metrics suggests that impression quality is inherently subjective and context dependent rather than defined by a single objective standard. Conclusion AI-generated radiology impressions, particularly those produced by custom domain-specific models, can achieve quality and safety comparable to human-authored impressions. These findings support the use of AI as an adaptable drafting aid that complements radiologist judgment. Biological sciences/Cancer Health sciences/Health care Health sciences/Medical research Health sciences/Oncology Radiology informatics artificial intelligence large language models clinical communication medical text generation Full Text Additional Declarations Competing interest reported. S.P., N.S., J.A., A.B., and A.D.G. are employees of Rad AI, which developed the custom domain-specific impression generation model evaluated in this study. S.C. and L.B. have a contractual relationship with Rad AI. All other authors declare no competing interests. Supplementary Files NPJDMSupplementary.docx Cite Share Download PDF Status: Published Journal Publication published 04 Apr, 2026 Read the published version in npj Digital Medicine → Version 1 posted Editorial decision: Revision requested 26 Jan, 2026 Reviews received at journal 25 Jan, 2026 Reviewers agreed at journal 15 Jan, 2026 Reviewers agreed at journal 13 Jan, 2026 Reviewers agreed at journal 13 Jan, 2026 Reviews received at journal 12 Jan, 2026 Reviewers agreed at journal 12 Jan, 2026 Reviewers invited by journal 12 Jan, 2026 Editor assigned by journal 02 Jan, 2026 Submission checks completed at journal 02 Jan, 2026 First submitted to journal 29 Dec, 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-8476600","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":573896011,"identity":"3870c120-139d-431b-8e5d-31e727afdb4f","order_by":0,"name":"Sharang Phadke","email":"","orcid":"","institution":"Rad AI","correspondingAuthor":false,"prefix":"","firstName":"Sharang","middleName":"","lastName":"Phadke","suffix":""},{"id":573896016,"identity":"34475a28-7190-4027-92fc-96bed2d0fb59","order_by":1,"name":"Nivedita Suresh","email":"","orcid":"","institution":"Rad AI","correspondingAuthor":false,"prefix":"","firstName":"Nivedita","middleName":"","lastName":"Suresh","suffix":""},{"id":573896022,"identity":"854d8295-db13-4bf7-9857-a769f3e4f16f","order_by":2,"name":"Zachary Allen","email":"","orcid":"","institution":"Rad AI","correspondingAuthor":false,"prefix":"","firstName":"Zachary","middleName":"","lastName":"Allen","suffix":""},{"id":573896027,"identity":"1d315c53-b59d-4be2-b642-8bab9047da88","order_by":3,"name":"Anjali Balagopal","email":"","orcid":"","institution":"Rad AI","correspondingAuthor":false,"prefix":"","firstName":"Anjali","middleName":"","lastName":"Balagopal","suffix":""},{"id":573896031,"identity":"b8540b00-053e-4955-824d-cb116dc596de","order_by":4,"name":"Stephen Chan","email":"","orcid":"","institution":"Rad AI","correspondingAuthor":false,"prefix":"","firstName":"Stephen","middleName":"","lastName":"Chan","suffix":""},{"id":573896036,"identity":"0b2f7266-a026-4843-82cf-0bcfd6cd97f4","order_by":5,"name":"Anish Shah","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Anish","middleName":"","lastName":"Shah","suffix":""},{"id":573896040,"identity":"bf2cb810-6262-4705-835a-a88725732bc7","order_by":6,"name":"Megan Winter","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Megan","middleName":"","lastName":"Winter","suffix":""},{"id":573896043,"identity":"00a0079e-491b-428e-8855-6f1b3f324a6c","order_by":7,"name":"Cesar Lam","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Cesar","middleName":"","lastName":"Lam","suffix":""},{"id":573896046,"identity":"769462d1-78d9-4700-8863-3621b54ffe00","order_by":8,"name":"Trevor Rose","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Trevor","middleName":"","lastName":"Rose","suffix":""},{"id":573896049,"identity":"858311cb-9912-4919-8338-aad71c74371f","order_by":9,"name":"Cyrillo Araujo","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Cyrillo","middleName":"","lastName":"Araujo","suffix":""},{"id":573896052,"identity":"b56e8dd2-ba94-459d-9cab-8e6b13a1362e","order_by":10,"name":"Abraham Ahmed","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Abraham","middleName":"","lastName":"Ahmed","suffix":""},{"id":573896055,"identity":"2f31daad-bfc4-4c73-a787-f3c13157fb9f","order_by":11,"name":"Iman Imanirad","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Iman","middleName":"","lastName":"Imanirad","suffix":""},{"id":573896060,"identity":"fc8e17d5-fc63-47b2-a3ff-59bf9f01e0e0","order_by":12,"name":"Lincoln Berland","email":"","orcid":"","institution":"Rad AI","correspondingAuthor":false,"prefix":"","firstName":"Lincoln","middleName":"","lastName":"Berland","suffix":""},{"id":573896063,"identity":"b8b8a9b8-fdf8-4b35-8f55-6ef30a574577","order_by":13,"name":"Andrew Del Gaizo","email":"data:image/png;base64,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","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":true,"prefix":"","firstName":"Andrew","middleName":"Del","lastName":"Gaizo","suffix":""}],"badges":[],"createdAt":"2025-12-30 02:23:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8476600/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8476600/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41746-026-02586-6","type":"published","date":"2026-04-04T15:59:13+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":100370643,"identity":"bf6b0eed-2e0a-4bcd-919f-ef77861ca1c4","added_by":"auto","created_at":"2026-01-16 08:07:07","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":417733,"visible":true,"origin":"","legend":"","description":"","filename":"NPJDMManuscriptSubmission1.docx","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/e2f0944a107052037b89e045.docx"},{"id":100242841,"identity":"6f1ac0e4-0861-47d2-ba2a-0a6cb54dcb40","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12846,"visible":true,"origin":"","legend":"","description":"","filename":"d45cfa9a456e4e478104b20d711129cd.json","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/67225e0e42dc8c8a023a1797.json"},{"id":100242842,"identity":"ee99c48f-5c2c-4248-9ec0-f9058e202784","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10766,"visible":true,"origin":"","legend":"","description":"","filename":"NPJDMSupplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/bd556afda7426f933096aafa.docx"},{"id":100242845,"identity":"9e6b52df-4c12-48f3-8d3f-03375569a9f0","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98637,"visible":true,"origin":"","legend":"","description":"","filename":"d45cfa9a456e4e478104b20d711129cd1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/b580664091e5894654f166b8.xml"},{"id":100242847,"identity":"3758af08-c7d2-478b-9b17-1e4769dcb5c6","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54111,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/91d8149081c5abb1420134a8.jpeg"},{"id":100371922,"identity":"c23715ef-4cb2-4b47-b55a-35cbb62190e5","added_by":"auto","created_at":"2026-01-16 08:11:14","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":70164,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/f8439bac57a972d0f9a7aa5b.jpeg"},{"id":100242844,"identity":"7c45c416-df46-4ef8-801e-3e21123472b2","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16823,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/e9850a6cfcbe47b7ab39ba88.png"},{"id":100242839,"identity":"6899f34f-37a1-41f0-af4d-db628c0bd738","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19222,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/55135bd8194f295f6f05e27c.png"},{"id":100242848,"identity":"f57a59aa-b280-49c0-b5c7-223df99632a1","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96415,"visible":true,"origin":"","legend":"","description":"","filename":"d45cfa9a456e4e478104b20d711129cd1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/befd217672f854415b60f787.xml"},{"id":100370444,"identity":"4494dede-49fc-4946-a2dd-b5949d42bb2f","added_by":"auto","created_at":"2026-01-16 08:05:49","extension":"html","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118155,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/89e5914395d483bffd7da22d.html"},{"id":106344975,"identity":"747ca2ca-ccb5-4649-9d50-f1a1b07eba8a","added_by":"auto","created_at":"2026-04-07 16:17:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":633606,"visible":true,"origin":"","legend":"","description":"","filename":"NPJDMManuscriptSubmission1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1_covered_450d55d7-fbea-43d7-8baf-bf6cc9ec9899.pdf"},{"id":100242838,"identity":"c252ca29-7451-4074-8e0c-175db72ef2a2","added_by":"auto","created_at":"2026-01-14 13:43:08","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":10766,"visible":true,"origin":"","legend":"","description":"","filename":"NPJDMSupplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-8476600/v1/44b2d888475ed4f65ebb1906.docx"}],"financialInterests":"Competing interest reported. S.P., N.S., J.A., A.B., and A.D.G. are employees of Rad AI, which developed the custom domain-specific impression generation model evaluated in this study. S.C. and L.B. have a contractual relationship with Rad AI. All other authors declare no competing interests.","formattedTitle":"Comparison of AI-Generated Radiology Impressions: A Multi-Stakeholder Evaluation","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":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Radiology informatics, artificial intelligence, large language models, clinical communication, medical text generation","lastPublishedDoi":"10.21203/rs.3.rs-8476600/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8476600/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate the quality, safety, and clinical utility of AI-generated radiology impressions compared with human-authored impressions across multiple clinical stakeholder groups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMaterials \u0026amp; Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA retrospective, blinded evaluation was conducted using 200 oncologic computed-tomography reports from a U.S. academic cancer center. Three impression types were assessed for each report: original radiologist-authored impressions, impressions generated by a custom domain-specific AI model fine-tuned on institutional data, and impressions generated by a general-purpose large language model. Original authoring radiologists, independent radiologists, and oncologists evaluated impressions using structured Likert-scale metrics assessing completeness, correctness, conciseness, clarity, clinical utility, and potential patient harm. Pairwise comparisons were performed using Wilcoxon signed-rank and two-proportion z-tests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCustom model AI impressions demonstrated near parity with human-authored impressions across most quality metrics. Original radiologists rated their own impressions as slightly more complete, while independent radiologists showed no significant differences between original and custom model impressions. Generic model impressions were longer, rated as more complete but significantly less concise. Patient harm ratings were uniformly low. Radiologists preferred original and custom model impressions over generic model impressions, whereas oncologists showed no significant preference.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscussion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEvaluation outcomes varied by stakeholder group, highlighting differing priorities between radiologists and oncologists. Low inter-rater agreement across several quality metrics suggests that impression quality is inherently subjective and context dependent rather than defined by a single objective standard.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAI-generated radiology impressions, particularly those produced by custom domain-specific models, can achieve quality and safety comparable to human-authored impressions. These findings support the use of AI as an adaptable drafting aid that complements radiologist judgment.\u003c/p\u003e","manuscriptTitle":"Comparison of AI-Generated Radiology Impressions: A Multi-Stakeholder Evaluation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-14 13:43:03","doi":"10.21203/rs.3.rs-8476600/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-27T01:16:53+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-25T12:10:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"87829652574017702987661515355113599072","date":"2026-01-15T08:23:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"98047858058243879102596764535718579766","date":"2026-01-13T09:38:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"338468345986801536723283444961096712157","date":"2026-01-13T05:28:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-13T02:50:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"99773548497550534624720903134909398020","date":"2026-01-12T23:42:25+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-12T23:27:48+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-02T19:11:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-02T11:32:33+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Digital Medicine","date":"2025-12-30T02:14:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6f63dc88-fed1-4b65-ab9d-ad9e1c552e54","owner":[],"postedDate":"January 14th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":61063867,"name":"Biological sciences/Cancer"},{"id":61063868,"name":"Health sciences/Health care"},{"id":61063869,"name":"Health sciences/Medical research"},{"id":61063870,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2026-04-07T16:15:38+00:00","versionOfRecord":{"articleIdentity":"rs-8476600","link":"https://doi.org/10.1038/s41746-026-02586-6","journal":{"identity":"npj-digital-medicine","isVorOnly":false,"title":"npj Digital Medicine"},"publishedOn":"2026-04-04 15:59:13","publishedOnDateReadable":"April 4th, 2026"},"versionCreatedAt":"2026-01-14 13:43:03","video":"","vorDoi":"10.1038/s41746-026-02586-6","vorDoiUrl":"https://doi.org/10.1038/s41746-026-02586-6","workflowStages":[]},"version":"v1","identity":"rs-8476600","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8476600","identity":"rs-8476600","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}