Title: Privacy preserving attribute-focused anonymization scheme for healthcare data publishing

Abstract

The advancements in industry 4.0 brought tremendous improvements in the healthcare sector such as better quality of treatment, enhanced communication, remote monitoring, reduced cost, and so on. Sharing healthcare data with healthcare providers is crucial in harnessing the benefits of the improvements. Healthcare big data generally contains sensitive information about individuals. Hence, sharing such data is challenging due to various security and privacy issues. According to the privacy regulations and ethical requirements, it is essential to preserve the privacy of the patients before sharing the data for medical research. In this paper, we propose an attribute-focused privacy preserving data publication scheme. The proposed scheme is two-fold, it comprises of a fixed-interval approach to protect numerical attributes and an improved l-diverse slicing approach to protect the categorical and sensitive attributes. The proposed scheme is effective in thwarting privacy attacks such as identity disclosure, attribute disclosure, and membership disclosure even when the adversary possesses concise background knowledge. In the fixed-interval approach, the original values of the healthcare data are replaced with an equivalent estimated value. The improved l-diverse slicing approach protects the data from various privacy risks. Therefore, the proposed scheme ensures both privacy and data utility of the published healthcare data. Extensive experiments with real datasets are conducted to evaluate the performance of the proposed scheme. Experimental analyses show that the proposed scheme is efficient in preserving the privacy and data utility with less computational complexity.