A cost effective clustering based anonymization approach for storing PHR's in cloud

As there is an increasing need to share the medical information for public health research, enormous amount of Personal Health Records (PHR's) are periodically collected and shared between two or many sources for research purpose. Sharing medical information about an individual without revealing sensitive information is the biggest challenge. Privacy and security are the two biggest obstacles for this process. Since medical information is related to human subjects, it is essential to preserve the privacy of the patients and ensure security to the medical information stored in cloud. In this paper, privacy of the shared PHR's is preserved through data anonymization and encryption algorithm. PHR's can be anonymized using various techniques such as generalization, suppression, truncation, etc., This paper focuses to provide efficient analysis of the shared PHR's by the proposed Efficient K-Means Clustering (EKMC) algorithm and to reduce the cost of data storage by the proposed Data Aggregation and Deduplication (DAD) algorithm. The EKMC algorithm is efficient and consumes less time when compared to the traditional k-means clustering algorithm. A set of performance analysis showing the effectiveness of our approach using synthetic data sets is presented.