AI-DRIVEN ZERO TRUST SECURITY FRAMEWORK FOR PROTECTING ELECTRONIC HEALTH RECORDS IN MODERN HEALTHCARE SYSTEMS

Abstract

The rapid digitization of healthcare services has increased dependency on Electronic Health Records (EHRs), making them a prime target for cyberattacks and data breaches. Conventional perimeter-based security mechanisms fail to protect against insider threats, compromised credentials, and lateral movement within healthcare networks. This research introduces an AI-Driven Zero Trust Security Framework (ZTSF) designed specifically for EHR environments. The proposed model implements continuous verification based on user identity, device posture, behavioral analytics, and context-aware access policies. Artificial intelligence techniques, including anomaly detection and predictive threat modeling, dynamically evaluate access requests and detect deviations from normal activity in real time. Experimental evaluation demonstrates significant improvements in threat mitigation and authorization efficiency compared with static access control models. By integrating Zero Trust principles with machine learning capabilities, the framework enhances confidentiality, integrity, and regulatory compliance for patient health data. The results indicate that the ZTSF is scalable and suitable for hospitals, telemedicine systems, and distributed healthcare networks.