Privacy-Preserving Disease Prediction with Secure Data Deduplication on Untrusted Cloud Servers

In the era of digital healthcare, ensuring the privacy of sensitive health data while optimizing storage on cloud servers is paramount. This paper introduces a novel privacy-preserving framework that not only predicts diseases based on encrypted symptom data but also employs secure data deduplication techniques to significantly reduce storage requirements. Our approach utilizes homomorphic encryption to safeguard patient identities and a unique dual-level deduplication process that efficiently handles disease prescriptions. We demonstrate this through rigorous testing of the effectiveness of our system in maintaining privacy and optimizing cloud storage, setting a new standard for secure digital healthcare solutions.