Article Published In Vol.10 (Nov-Dec 2022)

Enhancing Cloud Security in Telemedicine using Zero Trust Architecture and CNN-LSTM for Data Protection

Author : Venkataramesh Induru, Priyadarshini Radhakrishnan, Vijai Anand Ramar and Purandhar. N

The rapid adoption of cloud-based telemedicine has enhanced healthcare accessibility but introduced significant security vulnerabilities, including data breaches, unauthorized access, and ransomware attacks. Traditional security models, which rely on perimeter-based defences, fail to address modern cyber threats due to their inherent trust assumptions. To mitigate these risks, this study integrates Zero Trust Architecture (ZTA) with deep learning-based anomaly detection using CNN-LSTM. ZTA enforces strict access control through multi-factor authentication (MFA), micro-segmentation, and continuous monitoring, reducing unauthorized access risks. Meanwhile, the CNN-LSTM model detects cyber threats by analysing spatial and temporal patterns in security logs, enabling real-time anomaly detection. Experimental results demonstrate that the proposed model significantly improves cloud security in telemedicine, achieving a 98.5% accuracy in threat detection. Compared to traditional security methods, which often fail to detect sophisticated cyber threats, this approach reduces unauthorized access attempts by over 90%, enhancing patient data protection. Furthermore, the system continuously learns and adapts to evolving threats, ensuring sustained security improvements over time. The results confirm that combining ZTA with deep learning enhances security, privacy, and compliance in cloud-based telemedicine, making it a viable solution for safeguarding sensitive healthcare data.

Keywords: Zero Trust Architecture, Convolutional Neural Network, long short-term memory, telemedicine security, anomaly detection, multi-factor authentication, cloud data protection, cyber threat detection, deep learning.