Evaluating Transparency of Predetermined Change Control Plans in FDA-Cleared Radiology AI Devices: A Systematic Scoping Review

Evaluating Transparency of Predetermined Change Control Plans in FDA-Cleared Radiology AI Devices: A Systematic Scoping 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 Evaluating Transparency of Predetermined Change Control Plans in FDA-Cleared Radiology AI Devices: A Systematic Scoping Review Ketan Dayma, Palak Patel, Kenneth Hildreth, Tamara Jamaspishvili This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9411603/v2 This work is licensed under a CC BY 4.0 License Status: Posted Version 2 posted You are reading this latest preprint version Show more versions Abstract Background Adoption of FDA-cleared AI/ML-enabled devices is rising rapidly in Radiology. By recently denying a petition to exempt computer-aided detection (CAD) devices from 510(k) review, the FDA reaffirmed Predetermined Change Control Plans (PCCPs) as the preferred framework for managing post-market modifications. Purpose To characterize PCCP adoption and documentation transparency among FDA-cleared radiology AI/ML-devices. Materials and methods Cross-sectional systematic scoping review (PRISMA-ScR; PROSPERO) using linked data from FDA AI/ML-enabled device databases through April 2026. PCCP documentation completeness scored using an 8-point rubric derived from FDA's final PCCP guidance (December 2024). Results Among 1,394 FDA-listed AI/ML- device submissions, 1080 (77.5%, 870 unique devices) were radiology submissions, nearly all cleared via 510(k). Across all FDA panels, 170 devices were PCCP-cleared; radiology led AI/ML-specific PCCP adoption at 91.2% (34/37 radiology PCCP devices). Of 34 PCCP-cleared radiology AI devices, 65% (22/34) cleared in 2025 alone following final FDA guidance. Discrepancies between FDA's public database and individual summaries required manual adjudication for 27% of devices (9/34). PCCP documentation scores ranged from 0–8 (mean 5), with most modifications focused on data retraining, compatibility expansion and algorithm optimization. Continuous monitoring of device performance and predefined drift triggers for re-training were absent from public summaries. Conclusion PCCP adoption is accelerating in Radiology, yet public lifecycle controls, particularly monitoring performance metrics and trigger thresholds, remain sparse. Standardized PCCP reporting of lifecycle controls must be required as a condition of PCCP authorization to enable systematic post-market monitoring, as this pathway scales. Artificial Intelligence and Machine Learning Medical Informatics Health Policy Artificial Intelligence Radiology Predetermined Change Control Plan (PCCP) Post-Market Surveillance Software as a Medical Device (SaMD) Figures Figure 1 Figure 2 Full Text Additional Declarations The authors declare no competing interests. Supplementary Files daymaeappendix.docx daymapccpApril.xlsx Cite Share Download PDF Status: Posted Version 2 posted You are reading this latest preprint version Show more versions 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-9411603","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":623394857,"identity":"edbcb4f8-b9a8-4a5e-8c7b-8c5ad95000c7","order_by":0,"name":"Ketan Dayma","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYDCCA2DCBl2EsJY0IMEM4xCn5TAJWviOnzHd8OPM+cTt/OcPfv5QwSDHdyMBvxbJMzlmN3tu3E7cOSOZWeLAGQZjSUJaDA7kmN3g+XA7ccMNZgaJg20MQAYhLeffmN388+Fc4obzh5l/HPzHUE9Yy40cs9s8Nw4kbjiQzCZxsIEhwYCgX248K7stcybZGOgXM4szxyQMZ555gF8L3/nkbTffHLOT3c5/8PGNihobeb7jBGxhYOAwgLgQwpMgpBwE2B8gaxkFo2AUjIJRgAkA6dFXDJKorfkAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-6412-8327","institution":"SUNY Upstate Medical University","correspondingAuthor":true,"prefix":"","firstName":"Ketan","middleName":"","lastName":"Dayma","suffix":""},{"id":623394858,"identity":"c032efde-f171-4fc2-b8ed-5dc8bfd321e5","order_by":1,"name":"Palak Patel","email":"","orcid":"","institution":"SUNY Upstate Medical University","correspondingAuthor":false,"prefix":"","firstName":"Palak","middleName":"","lastName":"Patel","suffix":""},{"id":623394859,"identity":"5f12e046-41b0-469e-bb75-e1a98dfd973e","order_by":2,"name":"Kenneth Hildreth","email":"","orcid":"","institution":"SUNY Upstate Medical University","correspondingAuthor":false,"prefix":"","firstName":"Kenneth","middleName":"","lastName":"Hildreth","suffix":""},{"id":623394860,"identity":"1c5ffbd5-c911-4c4b-9aa9-cdc6f43906f7","order_by":3,"name":"Tamara Jamaspishvili","email":"","orcid":"","institution":"SUNY Upstate Medical University","correspondingAuthor":false,"prefix":"","firstName":"Tamara","middleName":"","lastName":"Jamaspishvili","suffix":""}],"badges":[],"createdAt":"2026-04-14 07:05:19","currentVersionCode":2,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9411603/v2","doiUrl":"https://doi.org/10.21203/rs.3.rs-9411603/v2","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107566887,"identity":"d0b6e0bf-8204-4da1-8999-a2a5e83d9be1","added_by":"auto","created_at":"2026-04-22 17:09:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":78398,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFDA Regulatory Pathway Clearance Trends for AI/ML and Traditional Medical Devices (2015–2025).\u003c/strong\u003e A dual-axis area chart comparing the annual count of FDA clearances for AI/ML devices (blue series, right vertical axis) and traditional non-AI/ML devices (red series, left vertical axis) from 2015 to 2025. Data points indicate the annual clearance count and the corresponding percentage of total FDA clearances for that year. The number and percentage share of clearances for AI/ML devices show a steady and accelerating rise, increasing from 6 clearances (0.2% of total) in 2015 to 331 clearances (11.5% of total) in 2025. In contrast, clearances for traditional non-AI/ML devices remained relatively stagnant throughout the decade, with their share of total clearances declining.\u003c/p\u003e","description":"","filename":"Picture1jama.png","url":"https://assets-eu.researchsquare.com/files/rs-9411603/v2/2709f6844dbc5023ec0e9d0f.png"},{"id":107567003,"identity":"496bbd7b-2fe5-4f4c-94f2-515303d3c130","added_by":"auto","created_at":"2026-04-22 17:09:50","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":198605,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePCCP Pathway Growth: AI/ML vs. Non-AI/ML Devices\u003c/strong\u003e. Data labels represent absolute PCCP clearances with proportional adoption rates (PCCP clearances / total category clearances) in parentheses. Notably, despite lower absolute volumes, AI/ML proportional adoption (10.9%) outpaced traditional devices (3.5%) in 2025.\u003c/p\u003e","description":"","filename":"Picture2jama.png","url":"https://assets-eu.researchsquare.com/files/rs-9411603/v2/9610ace5bfa073364b85ae76.png"},{"id":107706589,"identity":"e75845c2-4650-4d70-a82e-990b48034ade","added_by":"auto","created_at":"2026-04-24 09:18:25","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":600394,"visible":true,"origin":"","legend":"","description":"","filename":"daymamanuscriptpccprai.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9411603/v2_covered_dae1cfae-771e-48ec-bb3c-5b6dd3143d66.pdf"},{"id":107566885,"identity":"d4a99e10-8ef1-42d6-9d9e-5f705344c2cc","added_by":"auto","created_at":"2026-04-22 17:09:18","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":421080,"visible":true,"origin":"","legend":"","description":"","filename":"daymaeappendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-9411603/v2/c5e02d36623f61a5734b0284.docx"},{"id":107566931,"identity":"54c28464-0d89-4e31-a313-0f060f729944","added_by":"auto","created_at":"2026-04-22 17:09:32","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2151450,"visible":true,"origin":"","legend":"","description":"","filename":"daymapccpApril.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9411603/v2/b88708acfc25c83264c87919.xlsx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"Evaluating Transparency of Predetermined Change Control Plans in FDA-Cleared Radiology AI Devices: A Systematic Scoping Review","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Artificial Intelligence, Radiology, Predetermined Change Control Plan (PCCP), Post-Market Surveillance, Software as a Medical Device (SaMD)","lastPublishedDoi":"10.21203/rs.3.rs-9411603/v2","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9411603/v2","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eAdoption of FDA-cleared AI/ML-enabled devices is rising rapidly in Radiology. By recently denying a petition to exempt computer-aided detection (CAD) devices from 510(k) review, the FDA reaffirmed Predetermined Change Control Plans (PCCPs) as the preferred framework for managing post-market modifications.\u003c/p\u003e\u003ch2\u003ePurpose\u003c/h2\u003e \u003cp\u003eTo characterize PCCP adoption and documentation transparency among FDA-cleared radiology AI/ML-devices.\u003c/p\u003e\u003ch2\u003eMaterials and methods\u003c/h2\u003e \u003cp\u003eCross-sectional systematic scoping review (PRISMA-ScR; PROSPERO) using linked data from FDA AI/ML-enabled device databases through April 2026. PCCP documentation completeness scored using an 8-point rubric derived from FDA's final PCCP guidance (December 2024).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAmong 1,394 FDA-listed AI/ML- device submissions, 1080 (77.5%, 870 unique devices) were radiology submissions, nearly all cleared via 510(k). Across all FDA panels, 170 devices were PCCP-cleared; radiology led AI/ML-specific PCCP adoption at 91.2% (34/37 radiology PCCP devices). Of 34 PCCP-cleared radiology AI devices, 65% (22/34) cleared in 2025 alone following final FDA guidance. Discrepancies between FDA's public database and individual summaries required manual adjudication for 27% of devices (9/34). PCCP documentation scores ranged from 0\u0026ndash;8 (mean 5), with most modifications focused on data retraining, compatibility expansion and algorithm optimization. Continuous monitoring of device performance and predefined drift triggers for re-training were absent from public summaries.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003ePCCP adoption is accelerating in Radiology, yet public lifecycle controls, particularly monitoring performance metrics and trigger thresholds, remain sparse. Standardized PCCP reporting of lifecycle controls must be required as a condition of PCCP authorization to enable systematic post-market monitoring, as this pathway scales.\u003c/p\u003e","manuscriptTitle":"Evaluating Transparency of Predetermined Change Control Plans in FDA-Cleared Radiology AI Devices: A Systematic Scoping Review","msid":"","msnumber":"","nonDraftVersions":[{"code":2,"date":"2026-04-22 17:08:13","doi":"10.21203/rs.3.rs-9411603/v2","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}},{"code":1,"date":"2026-04-15 07:00:14","doi":"10.21203/rs.3.rs-9411603/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b0726150-1afc-4a73-a8eb-2e41d00f22b1","owner":[],"postedDate":"April 22nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66343284,"name":"Artificial Intelligence and Machine Learning"},{"id":66343285,"name":"Medical Informatics"},{"id":66343286,"name":"Health Policy"}],"tags":[],"updatedAt":"2026-04-15T07:00:15+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-22 17:08:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v2","identity":"rs-9411603","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9411603","identity":"rs-9411603","version":["v2"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}