Journal article
MediChainAudit: Encrypted Medical Data Auditing on Blockchain with Fuzzy Deduplication and Secure Sharing
IEEE transactions on dependable and secure computing, Vol.First online, pp.1-17
17/08/2026
Metrics
1 Record Views
Abstract
Cloud storage increasingly uses deduplication to re duce costs, while users require strong confidentiality and integrity for outsourced data—especially in healthcare, where sensitive, redundant data is frequently shared under strict privacy constraints. Although Message-Locked Encryption (MLE) supports encrypted deduplication, existing methods focus on exact duplicates and assume trusted auditors, making them unsuitable for medical use. We propose MediChainAudit, a blockchain-based system for medical clouds that integrates: (i) an enhanced encryption mode enabling fuzzy (similarity-aware) block-level dedu plication, (ii) a decentralized public audit framework removing the need for trusted third parties, and (iii) a secure key-exchange and sharing protocol. Using a hybrid storage model—on-chain integrity tags and off-chain ciphertext—the system minimizes blockchain overhead while supporting verifiable secure deletion. We formalize security goals including PRV-CDA confidentiality, integrity, ownership consistency, and deduplication correctness, providing proofs under standard cryptographic assumptions. To our knowledge, MediChainAudit is the first blockchain-based solution to integrate fuzzy deduplication auditing, secure sharing, and side-channel attack defense with proven PRV-CDA security.
Details
- Title
- MediChainAudit: Encrypted Medical Data Auditing on Blockchain with Fuzzy Deduplication and Secure Sharing
- Creators
- Jie Zhang - Tianjin UniversityXiaohong Li - Tianjin UniversityRuitao Feng - Southern Cross UniversityShanshan Xu - East China Normal UniversityYongyang Lv - Tianjin UniversityZhe Hou - Griffith UniversityGuangdong Bai - City University of Hong Kong
- Publication Details
- IEEE transactions on dependable and secure computing, Vol.First online, pp.1-17
- Publisher
- IEEE
- Identifiers
- 991013396950002368
- Copyright
- © 2026, IEEE.
- Academic Unit
- Faculty of Science and Engineering
- Language
- English
- Resource Type
- Journal article