A Comparative Study of Differential Privacy Mechanisms on a ZIKV Dataset in Brazil

• Daniel de Oliveira UFF
• Eduardo Rodrigues UFC
• Serafim Costa UFC
• Paulo Amora UFC
• Asley Caldas UFC
• Kary Ocaña LNCC
• Marco Horta Fiocruz
• Ana Maria de Filippis Fiocruz
• Vânia Vidal UFC
• Javam Machado UFC

DOI: https://doi.org/10.5753/sbbd.2019.8832

Abstract

In recent years, the Brazilian Government has fostered a series of initiatives to automate and optimize the SUS (Brazilian national health system), aiming at improving its efficiency. One initiative is the Laboratory Environment Manager (GAL). The GAL aims at providing management of laboratory routines and follow-up of the steps to perform the examinations. Also, the GAL allows for data extraction, which can be used by managers in the various governmental spheres. However, such data export may be unreliable and lead to severe risks of privacy violation as it shows the personal and sensitive data of individuals. Merely masking the identifying elements (name, social security number, etc.) or providing only aggregate results may not provide sufficient protection. In this scenario, more elaborate data privacy techniques, such as Differential Privacy (DP), are required. This paper presents a study comparing the application of different DP mechanisms to data extracted from the GAL. In particular, we used as a case study the data from suspicious cases of Zika Virus (ZIKV) in Brazil.

Keywords: Differential Privacy, ZIKV, Privacy Mechanisms

References

Dagher, G. G., Mohler, J., Milojkovic, M., and Marella, P. B. (2018). Ancile: Privacypreserving framework for access control and interoperability of electronic health records using blockchain technology. Sustainable Cities and Society, 39:283 – 297. DOI: https://doi.org/10.1016/j.scs.2018.02.014

Dwork, C., McSherry, F., Nissim, K., and Smith, A. (2006). Calibrating noise to sensitivity in private data analysis. In Halevi, S. and Rabin, T., editors, Theory of Cryptography, pages 265–284, Berlin, Heidelberg. Springer Berlin Heidelberg. DOI: https://doi.org/10.1007/11681878_14

Erlingsson, Ú., Korolova, A., and Pihur, V. (2014). RAPPOR: randomized aggregatable privacy-preserving ordinal response. CoRR, abs/1407.6981. DOI: https://doi.org/10.1145/2660267.2660348

Kifer, D. and Machanavajjhala, A. (2011). No free lunch in data privacy. In Proc. of the 2011 SIGMOD, SIGMOD ’11, pages 193–204, New York, NY, USA. ACM. DOI: https://doi.org/10.1145/1989323.1989345

Li, S., Bamidis, P. D., Konstantinidis, S. T., Traver, V., Car, J., and Zary, N. (2019). Setting priorities for EU healthcare workforce IT skills competence improvement. Health Inf. Journal, 25(1). DOI: https://doi.org/10.1177/1460458217704257

Nascimento, F., Vale, K. O., and Gorgônio, F. L. (2018). Um estudo comparative entre algoritmos de proteção da privacidade aplicado à bases de dados na área de saúde. In XXXIII SBBD, Rio de Janeiro, RJ, Brazil, August 25-26, 2018., pages 301–306.

Shishvan, O. R., Zois, D., and Soyata, T. (2018). Machine intelligence in healthcare and medical cyber physical systems: A survey. IEEE Access, 6:46419–46494. DOI: https://doi.org/10.1109/ACCESS.2018.2866049

Silva, A. B., Guedes, A., Síndico, S., Vieira, E., and de Andrade Filha, I. (2019). Registro eletrônico de saúde em hospital de alta complexidade: um relato sobre o processo de implementação na perspectiva da telessaúde. Ciência & Saúde Coletiva, 24:1133–1142. DOI: https://doi.org/10.1590/1413-81232018243.05982017

Warner, S. L. (1965). Randomized response: A survey technique for eliminating evasive answer bias. Journal of the American Statistical Association, 60(309):63–69. DOI: https://doi.org/10.2307/2283137