Intitulée : Cadre d'intelligence artificielle pour la prévention et la détection des attaques dans l'environnement IoT
Présenté par : Héla MAKINA
Membres de jury:
|
M. Taoufik AGUILI |
Professeur, Univ. of Tunis el Manar, Tunisia |
Président |
|
M. Khaled GRAYAA |
Professeur, Univ. of Carthage, Tunisia |
Rapporteur |
|
M. Bassem BEN SALAH |
Maître assistant, Univ. of Carthage, Tunisia |
Rapporteur |
|
Mme Soumaya HAMMOUDA |
Professeur, Univ. of Carthage, Tunisia |
Examinatrice |
|
Mme Asma BEN LETAIFA |
Maître assistante, Univ. of Carthage, Tunisia |
Directrice de thèse |
Résumé:
The management of healthcare data presents multifaceted challenges due to the sensitive nature of medical records, the strict requirements for timely access, the involvement of multiple stakeholders, and the regulatory obligations governing such information.
This applied research thesis aims to design and implement an efficient, secure, patient-centric, and regulation-compliant framework for Electronic Health Records (EHRs) management through the integration of decentralized technologies.
Following a comprehensive review of existing EHR solutions, we propose a novel multi-layer architecture that combines decentralized identity management using Decentralized Identifiers (DIDs), fine-grained access control through Verifiable Credentials (VCs), and secure data storage supported by blockchain technology and smart contracts.
Two architectural variants were developed and evaluated. The first is a fully decentralized model integrating SSI, blockchain, and IPFS, which guarantees data ownership, integrity, transparency, and resilience against single points of failure. The second is a hybrid model combining SSI, blockchain, and MongoDB to enhance performance, reduce latency, and better meet regulatory requirements, particularly the right to data erasure. A comparative evaluation reveals the inherent trade-offs between the two architectures: while the decentralized model excels in trustlessness and resilience, the hybrid model offers better transaction times, scalability, and regulatory compliance, highlighting the tension between ideal security properties and practical deployment needs.
Overall, this research demonstrates the practical feasibility of integrating blockchain and SSI into healthcare ecosystems, offering a scalable, secure, and patient-empowered solution for managing sensitive medical data.
Mots clés:
eHealth, Electronic Health Records (EHRs), Blockchain Technology, Self-Sovereign Identity (SSI), Decentralized Identifiers (DID), Verifiable Credentials (VC), Access Control, Privacy and Security
Maintenant, allez pousser vos propres limites et réussir!