Digital Twin Cloud Platform for Healthcare Monitoring Based on Privacy Similarity Queries

Hao Chen¹Jiaxing Xiu^2*

• ¹La Consolacion University Philippines, Malolos, Philippines
• ²Xi'an Jiaotong University, Xi'an, China

Efficient health monitoring while safeguarding patient privacy remains a critical challenge in modern medical surveillance. This study proposes a digital twin cloud platform for medical monitoring, leveraging privacy-preserving similarity queries to address the dual challenges of privacy protection and monitoring efficiency. The platform integrates visual Internet of Thing (IoT) simulation modeling and eventdriven virtual-reality mapping technologies to enable real-time, bidirectional data coupling between physical medical devices and virtual twin models, ensuring synchronous updates and accurate monitoring. At the algorithmic level, this study introduces a privacy-preserving K Nearest Neighbor (KNN) query method, which dynamically generates terminal-specific encryption functions for data and queries. A bucket-based data encoding scheme is employed to ensure cipher text in distinguishability, while a bucket distance-based similarity metric allows direct similarity comparisons in encrypted environments, preserving data utility without compromising privacy. Experimental results demonstrate the platform's superiority in query precision, accuracy, time efficiency, and privacy protection strength compared to existing methods. The proposed solution not only validates the feasibility of privacypreserving medical monitoring but also offers an innovative approach to secure and efficient healthcare data processing.