[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84050-en":3,"doc-seo-84050-105":28,"detail-sidebar-cat-0-en-105":89},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":13,"seo_description":14,"update_tm":26,"read_time":27},84050,1374391975076,"Riley","https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051",8,"Research & Report","REAN Reconstruction-aware ECG Anonymization Based on Privacy–Utility Orthogonality","Shared electrocardiograms (ECG) act as biometric fingerprints that can enable re-identification and expose personal information, creating a privacy–utility trade-off for signal anonymization. Existing approaches often either degrade clinical utility or leave biometric recovery feasible. REAN (REconstruction-aware ECG Anonymizer) anonymizes raw ECG via a 1-D U-Net trained with frozen privacy and utility classifiers. Its privacy and utility gradients are nearly orthogonal (~93.8°), yielding stronger privacy–utility balance on four PhysioNet datasets.","REAN: Reconstruction-aware ECG Anonymization Based on Privacy–Utility Orthogonality  \n1st Taerin Ki Chung-Ang University Seoul, South Korea [rlxofls@cau.ac.kr](rlxofls@cau.ac.kr)  \n2nd Sunghwan Park Chung-Ang University Seoul, South Korea [tjdghks994@cau.ac.kr](tjdghks994@cau.ac.kr)  \n3rd Junyoung Park Chung-Ang University Seoul, South Korea [june295921@cau.ac.kr](june295921@cau.ac.kr)  \n4th Jaewoo Lee Chung-Ang University Seoul, South Korea [jaewoolee@cau.ac.kr](jaewoolee@cau.ac.kr)  \narXiv :2607 .06037v 1 [ cs .CR] 7 Jul 2026  \nAbstract—A shared electrocardiogram (ECG) is itself a biometric fingerprint that can re-identify a patient and reveal personal information. Recent ECG anonymizers transform the signal before sharing to reduce privacy leakage. However, existing methods still face a privacy–utility trade-off, in which preserving privacy often compromises utility while preserving utility reveals personal information. We propose REAN (REconstruction-aware ECGANonymizer), a raw ECG signal anonymizer, to address this privacy–utility trade-off. REAN reconstructs the signal using a 1-D U-Net trained with losses from frozen privacy and utility classifiers to reduce privacy leakage while preserving utility. The privacy and utility gradients are near-orthogonal (≈93.8◦ ), so reducing privacy leakage leaves utility almost unchanged. On four public PhysioNet databases, REAN achieves the strongest privacy–utility balance among raw ECG signal baselines. It drives re-identification to chance (0.96→0.00), keeps arrhythmia macroAUROC at the clean level (Clean 0.9982 vs. REAN 0.9991), and maintains re-identification protection under unseen privacyclassifier architectures.  \nIndex Terms—ECG Privacy, Anonymization, Privacy–Utility Trade-off, Signal Reconstruction, Gradient Orthogonality  \nI. INTRODUCTION  \nElectrocardiograms (ECGs) are widely shared for diagnosing arrhythmia, ischemia, and conduction disorders, yet the same waveform is a biometric fingerprint that reveals a patient’s identity, gender, and age [1], [2] . Removing identifiers does not protect the patient, because the waveform itself is aquasi-identifier that enables re-identification from partial side knowledge [3], [4] . An effective defense must therefore transform the signal itself rather than its metadata.  \nTransforming the signal faces a privacy–utility trade-off between hiding biometric information and preserving diagnosis. Recent signal-level methods either degrade diagnosis or leave biometric information recoverable [5]–[8], because the two are entangled in the same morphology. This trade-off is usually treated as the unavoidable price of anonymization. Fig. 1 shows this trade-off across existing anonymizers. We instead ask the central question of this paper. Can biometric information be removed from an ECG without damaging the morphology that diagnosis depends on?  \nWe answer this question by observing that the utility and privacy directions are nearly orthogonal in ECG signal space. Measured as input gradients, the two directions meet at about 89.8◦ (Section II-B) . Biometric information can therefore be  \nFig. 1. The privacy–utility trade-off across ECG anonymizers. The lock markshow much identity and personal-information leakage the defense removes, and the hand marks the arrhythmia-diagnosis accuracy retained after anonymization, combined so that leaving any single attribute exposed keeps the score low. Only REAN scores high (green) on both axes.  \nsuppressed along the privacy direction with almost no effect on diagnosis.  \nWe propose REAN (REconstruction-aware ECGANonymizer), a 1-D U-Net that exploits this geometry and anonymizes an ECG in a single forward pass. REAN trains one objective that preserves diagnosis, protects privacy, and limits distortion, using a frozen diagnostic classifier and three frozen biometric classifiers as training signals. The utility and privacy terms optimize together with little conflict, because their gradients are orthogonal. A ","cbCaienBxUmVOESC","https://ap.wps.com/l/cbCaienBxUmVOESC","pdf",773158,1,"English","en",105,"# Abstract\n# Introduction\n## Motivation and privacy–utility trade-off\n## Key idea: orthogonality in ECG space\n## Contributions\n# Preliminary\n## ECG signal fundamentals\n## Orthogonality of utility and privacy directions","[{\"question\":\"Why can sharing ECG signals still compromise patient privacy even without metadata?\",\"answer\":\"Because the ECG waveform itself functions as a quasi-identifier, enabling re-identification from side information such as morphology-related cues.\"},{\"question\":\"What is the core idea behind REAN’s anonymization method?\",\"answer\":\"REAN leverages near-orthogonality between privacy and utility directions in ECG signal space, so moving the signal along the privacy direction suppresses biometric leakage while minimally affecting diagnostic morphology.\"},{\"question\":\"How does REAN achieve the privacy–utility balance in experiments?\",\"answer\":\"On four public PhysioNet databases, REAN drives re-identification from about 0.96 to chance while keeping arrhythmia macro-AUROC close to the clean level, and remains robust to unseen privacy-classifier architectures.\"}]",1784192247,20,{"code":4,"msg":29,"data":30},"ok",{"site_id":23,"language":22,"slug":31,"title":13,"keywords":32,"description":14,"schema_data":33,"social_meta":84,"head_meta":86,"extra_data":88,"updated_unix":26},"rean-reconstruction-aware-ecg-anonymization-based-on-privacyutility-orthogonality","",{"@graph":34,"@context":83},[35,52,66],{"@type":36,"itemListElement":37},"BreadcrumbList",[38,42,46,49],{"item":39,"name":40,"@type":41,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":43,"name":44,"@type":41,"position":45},"https://docshare.wps.com/document/","Document",2,{"item":47,"name":12,"@type":41,"position":48},"https://docshare.wps.com/document/research-report/",3,{"item":50,"name":13,"@type":41,"position":51},"https://docshare.wps.com/document/rean-reconstruction-aware-ecg-anonymization-based-on-privacyutility-orthogonality/84050/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":39,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":63,"interactionType":64,"userInteractionCount":4},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69,75,79],{"name":70,"@type":71,"acceptedAnswer":72},"Why can sharing ECG signals still compromise patient privacy even without metadata?","Question",{"text":73,"@type":74},"Because the ECG waveform itself functions as a quasi-identifier, enabling re-identification from side information such as morphology-related cues.","Answer",{"name":76,"@type":71,"acceptedAnswer":77},"What is the core idea behind REAN’s anonymization method?",{"text":78,"@type":74},"REAN leverages near-orthogonality between privacy and utility directions in ECG signal space, so moving the signal along the privacy direction suppresses biometric leakage while minimally affecting diagnostic morphology.",{"name":80,"@type":71,"acceptedAnswer":81},"How does REAN achieve the privacy–utility balance in experiments?",{"text":82,"@type":74},"On four public PhysioNet databases, REAN drives re-identification from about 0.96 to chance while keeping arrhythmia macro-AUROC close to the clean level, and remains robust to unseen privacy-classifier architectures.","https://schema.org",{"og:url":50,"og:type":85,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":87,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":90},[91,95,99,103,108,113,118,121,125,128,132],{"id":20,"doc_module":4,"doc_module_name":44,"category_name":92,"show_sort_weight":93,"slug":94},"Story & Novel",90,"story-novel",{"id":45,"doc_module":4,"doc_module_name":44,"category_name":96,"show_sort_weight":97,"slug":98},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":44,"category_name":100,"show_sort_weight":101,"slug":102},"Exam",70,"exam",{"id":104,"doc_module":4,"doc_module_name":44,"category_name":105,"show_sort_weight":106,"slug":107},5,"Comic",60,"comic",{"id":109,"doc_module":4,"doc_module_name":44,"category_name":110,"show_sort_weight":111,"slug":112},6,"Technology",50,"technology",{"id":114,"doc_module":4,"doc_module_name":44,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":44,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":44,"category_name":123,"show_sort_weight":27,"slug":124},9,"Religion & Spirituality","religion-spirituality",{"id":27,"doc_module":4,"doc_module_name":44,"category_name":126,"show_sort_weight":27,"slug":127},"World Cup","world-cup",{"id":129,"doc_module":4,"doc_module_name":44,"category_name":130,"show_sort_weight":129,"slug":131},10,"Lifestyle","lifestyle",{"id":133,"doc_module":4,"doc_module_name":44,"category_name":134,"show_sort_weight":104,"slug":135},19,"General","general"]