{"ID":2856153,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.11073","arxiv_id":"2510.11073","title":"ROFI: A Deep Learning-Based Ophthalmic Sign-Preserving and Reversible Patient Face Anonymizer","abstract":"Patient face images provide a convenient mean for evaluating eye diseases, while also raising privacy concerns. Here, we introduce ROFI, a deep learning-based privacy protection framework for ophthalmology. Using weakly supervised learning and neural identity translation, ROFI anonymizes facial features while retaining disease features (over 98\\% accuracy, $κ\u003e 0.90$). It achieves 100\\% diagnostic sensitivity and high agreement ($κ\u003e 0.90$) across eleven eye diseases in three cohorts, anonymizing over 95\\% of images. ROFI works with AI systems, maintaining original diagnoses ($κ\u003e 0.80$), and supports secure image reversal (over 98\\% similarity), enabling audits and long-term care. These results show ROFI's effectiveness of protecting patient privacy in the digital medicine era.","short_abstract":"Patient face images provide a convenient mean for evaluating eye diseases, while also raising privacy concerns. Here, we introduce ROFI, a deep learning-based privacy protection framework for ophthalmology. Using weakly supervised learning and neural identity translation, ROFI anonymizes facial features while retaining...","url_abs":"https://arxiv.org/abs/2510.11073","url_pdf":"https://arxiv.org/pdf/2510.11073v1","authors":"[\"Yuan Tian\",\"Min Zhou\",\"Yitong Chen\",\"Fang Li\",\"Lingzi Qi\",\"Shuo Wang\",\"Xieyang Xu\",\"Yu Yu\",\"Shiqiong Xu\",\"Chaoyu Lei\",\"Yankai Jiang\",\"Rongzhao Zhang\",\"Jia Tan\",\"Li Wu\",\"Hong Chen\",\"Xiaowei Liu\",\"Wei Lu\",\"Lin Li\",\"Huifang Zhou\",\"Xuefei Song\",\"Guangtao Zhai\",\"Xianqun Fan\"]","published":"2025-10-13T07:12:23Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}