Automated De-Identification, Consistent Obfuscation, and Regulatory Grade Validation of 2 Billion Patient Notes

Automated De-Identification, Consistent Obfuscation, and Regulatory Grade Validation of 2 Billion Patient Notes | 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 Automated De-Identification, Consistent Obfuscation, and Regulatory Grade Validation of 2 Billion Patient Notes Veysel Kocaman, Lindsay Mico, Mustafa Aytug Kaya, Nadaa Taiyab, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6867162/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Rich Large, diverse collections of anonymous patient data—including text, numbers, and images— are essential to advancing a broad range of causes, from clinical decision support and real-world evidence to population health and hospital operations. This study presents a novel system used to automatically de-identify unstructured clinical text from 2 billion patient notes, using consistent obfuscation and tokenization to link them into a unified longitudinal dataset. To the best of our knowledge, this is the first such system to be externally certified for regulatory-grade accuracy on real-world data at this scale. The system is based on proprietary medical language models and the modified Spark NLP - a distributed computing NLP framework for efficient execution on large clusters of commodity hardware. It satisfies the Expert Determination de-identification criteria under HIPAA ( Health Insurance Portability and Accountability Act) Privacy Rules, establishing a baseline requirement of <5% PHI prevalence both in aggregate and per record. It achieves 99% Protected Health Information (PHI) obfuscation, and achieves 100% masking or shifting of target data fields. This level of accuracy surpasses even that of a triple manual review by 3 human annotators. Obfuscation adds another layer of protection by rendering PHI elements indistinguishable from missed elements. Name changes, date shifting, and tokenizing identifiers are done consistently across documents about the same patient. Equity analysis was performed to ensure the system is not biased across demographic groups for gender, age, ethnicity, and state. Finally, an independent audit including adversarial testing on 790 randomly selected patients was performed, in which a dedicated “red team” working for 3 months was not able to re-identify any of the patients. Physical sciences/Mathematics and computing/Applied mathematics Physical sciences/Mathematics and computing/Computer science Physical sciences/Mathematics and computing/Information technology Physical sciences/Mathematics and computing/Scientific data Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-6867162","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":510279544,"identity":"9ae5ff77-8b8e-4e42-97e5-dcd3302a054d","order_by":0,"name":"Veysel Kocaman","email":"","orcid":"","institution":"John Snow Labs","correspondingAuthor":false,"prefix":"","firstName":"Veysel","middleName":"","lastName":"Kocaman","suffix":""},{"id":510279545,"identity":"822f7c45-b27e-48bf-a925-af1139a236df","order_by":1,"name":"Lindsay Mico","email":"","orcid":"","institution":"Providence Health","correspondingAuthor":false,"prefix":"","firstName":"Lindsay","middleName":"","lastName":"Mico","suffix":""},{"id":510279546,"identity":"77a28861-6983-45f5-b281-7c3c7591f62b","order_by":2,"name":"Mustafa Aytug Kaya","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAtElEQVRIiWNgGAWjYDACHhBRYYPEIU7LmTRStTC2HSZBi27P4WcSP9jO262dkcD44G0bEVrMzraZSfbw3E7ediOB2XAuUVrOM5hJ8EjcTja7kcAmzUucFvZvkn8MzoG0sP8mTsvZHjNpnoQDdiBbmInTcuZMsbXMgeQEszMPmyXnnCNKS/rGm2//2dmbHU8++OFNGRFagIBFAkgkNjAwNhCnHgiYPwAJe6KVj4JRMApGwcgDABUXORKjejonAAAAAElFTkSuQmCC","orcid":"","institution":"George Mason University","correspondingAuthor":true,"prefix":"","firstName":"Mustafa","middleName":"Aytug","lastName":"Kaya","suffix":""},{"id":510279547,"identity":"ba7f9961-cb67-4663-be6b-7b8ece82037b","order_by":3,"name":"Nadaa Taiyab","email":"","orcid":"","institution":"Providence Health","correspondingAuthor":false,"prefix":"","firstName":"Nadaa","middleName":"","lastName":"Taiyab","suffix":""},{"id":510279548,"identity":"af697ddd-379c-403b-98c9-fed03e1ca068","order_by":4,"name":"David Talby","email":"","orcid":"","institution":"John Snow Labs","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Talby","suffix":""},{"id":510279549,"identity":"3c4bba7e-9b93-466b-972e-1548b544b475","order_by":5,"name":"Tae Surh","email":"","orcid":"","institution":"Providence Health","correspondingAuthor":false,"prefix":"","firstName":"Tae","middleName":"","lastName":"Surh","suffix":""},{"id":510279550,"identity":"9c037a50-8969-43c1-97c7-75193d75e27d","order_by":6,"name":"Yuqing Guo","email":"","orcid":"","institution":"Providence Health","correspondingAuthor":false,"prefix":"","firstName":"Yuqing","middleName":"","lastName":"Guo","suffix":""},{"id":510279551,"identity":"8d6aa191-459b-4bf5-8f0e-26bc0dc6c732","order_by":7,"name":"Vivek Tomer","email":"","orcid":"","institution":"Providence Health","correspondingAuthor":false,"prefix":"","firstName":"Vivek","middleName":"","lastName":"Tomer","suffix":""},{"id":510279552,"identity":"2523c834-66c5-4dbb-ad38-27c56a015acd","order_by":8,"name":"Robert Kramer","email":"","orcid":"","institution":"Providence Health","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"","lastName":"Kramer","suffix":""}],"badges":[],"createdAt":"2025-06-11 02:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6867162/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6867162/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101676814,"identity":"adea2088-ee1a-40d0-a7b3-ffe795f4539c","added_by":"auto","created_at":"2026-02-02 13:41:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":456682,"visible":true,"origin":"","legend":"","description":"","filename":"mtnwgzqsgpvyzktvjsvvhsmfmpwwdkqc.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6867162/v1_covered_04db574a-1769-4294-ab16-555c043eb171.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Automated De-Identification, Consistent Obfuscation, and Regulatory Grade Validation of 2 Billion Patient Notes","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-6867162/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6867162/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRich Large, diverse collections of anonymous patient data—including text, numbers, and images— are essential to advancing a broad range of causes, from clinical decision support and real-world evidence to population health and hospital operations. This study presents a novel system used to automatically de-identify unstructured clinical text from 2 billion patient notes, using consistent obfuscation and tokenization to link them into a unified longitudinal dataset. To the best of our knowledge, this is the first such system to be externally certified for regulatory-grade accuracy on real-world data at this scale. The system is based on proprietary medical language models and the modified Spark NLP - a distributed computing NLP framework for efficient execution on large clusters of commodity hardware. It satisfies the Expert Determination de-identification criteria under HIPAA ( Health Insurance Portability and Accountability Act) Privacy Rules, establishing a baseline requirement of \u0026lt;5% PHI prevalence both in aggregate and per record. It achieves 99% Protected Health Information (PHI) obfuscation, and achieves 100% masking or shifting of target data fields. This level of accuracy surpasses even that of a triple manual review by 3 human annotators.\u003c/p\u003e\n\u003cp\u003eObfuscation adds another layer of protection by rendering PHI elements indistinguishable from missed elements. Name changes, date shifting, and tokenizing identifiers are done consistently across documents about the same patient. Equity analysis was performed to ensure the system is not biased across demographic groups for gender, age, ethnicity, and state. Finally, an independent audit including adversarial testing on 790 randomly selected patients was performed, in which a dedicated “red team” working for 3 months was not able to re-identify any of the patients.\u003c/p\u003e","manuscriptTitle":"Automated De-Identification, Consistent Obfuscation, and Regulatory Grade Validation of 2 Billion Patient Notes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-05 18:30:01","doi":"10.21203/rs.3.rs-6867162/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":"4c7b4866-904a-4eef-8b41-cd4edd043c1b","owner":[],"postedDate":"September 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":54201532,"name":"Physical sciences/Mathematics and computing/Applied mathematics"},{"id":54201533,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":54201534,"name":"Physical sciences/Mathematics and computing/Information technology"},{"id":54201535,"name":"Physical sciences/Mathematics and computing/Scientific data"}],"tags":[],"updatedAt":"2026-02-02T13:40:09+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-05 18:30:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6867162","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6867162","identity":"rs-6867162","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}