De-Identification of Health Data in Big Data using a Novel Bio-Inspired Apoptosis Algorithm

Abstract

In the last years, with the emergence of new technologies in the image of big data, the privacy concerns had grown widely. However, big data means the dematerialization of the data. The classical security solutions are no longer efficient in this case. Nowadays, sharing the data is much easier as well as saying hello. The amount of shared data over the web keeps growing from day to another which creates a wide gap between the purpose of sharing data and the fact that these last contain sensitive information. For that, the researches turned their attention to new issues and domains in order to minimize this gap. In other way, they intended to ensure a good utility of data by preserving its meaning while hiding sensitive information to prevent identity disclosure. Many techniques had been used for that. Some of it is mathematical and other ones using data mining algorithms. This paper deals with the problem of hiding sensitive data in shared structured medical data using a new bio-inspired algorithm from the natural phenomena of apoptosis cells in human body.

Secure Emergency Data Sharing in Healthcare - A Data Warehouse View

“…Repository-wide batch de-identification, on-demand cohort-specific de-identification, on-demand de-identification of query results, de-identification with patient and provider identifiers, scientist-involved de-identification, patient-involved de-identification, physician-involved de-identification, and online de-identification by honest brokers. Rahmani et al [21] use a novel bio-inspired algorithm based on the natural phenomenon of apoptotic cells in the human body to solve the challenge of concealing sensitive clinical data in big DWs. De-identification can be accomplished in a variety of ways.…”