Retracted: Investigating the Feasibility of Encryption as a Risk Mitigation Strategy for Machine Learning Algorithms in Medical Applications

Abstract

In current years, the usage of system learning algorithms for clinical purposes has become more and more ubiquitous, and the privacy and protection of affected person records have emerged as more vital topics of debate. As such, it is vital to explore diverse threat mitigation techniques. In this summary, we inspect the possibility of the usage of encryption as a danger mitigation strategy for gadget-mastering algorithms in medical programs. We first overview the specific forms of encryption to be had, after which we examine the feasibility of the use of encryption to guard patient records in a system-learning context. We examine the overall performance effects of encrypting data before the application of a machine studying the set of rules, in addition to capacity vulnerable factors that encryption by me might not be able to shield towards. Sooner or later, we speak about the quality practices for enforcing and the usage of encryption for medical programs regarding device getting-to-know algorithms. The outcomes of this analysis show that even as encryption is a feasible threat mitigation approach, quality practices nonetheless want to be set up for its implementation.

Ähnliche Arbeiten