Design And Implementation Of Medical Image Desensitization System

Medical image data contains great medical value,but it also contains a large number of sensitive information of patients.If the original medical image data is released or shared without desensitization,the patient’s privacy will inevitably be leaked.Data anonymization technology deals with data sets through anonymous operations such as concealment and generalization,so that data tuples can not be associated with specific individuals,thus reducing the risk of privacy disclosure.Aiming at the problem of personalized privacy preservation in the field of data anonymization,this paper improves the PE(α,k)model and proposes an IPE(α,k)model,which can not only meet the individual privacy preservation needs,but also avoid similarity attacks.At the same time,a personalized anonymous algorithm based on clustering-merging-splitting is proposed,which can further reduce information loss and retain more data availability on the basis of implementing IPE(α,k)model.Based on the proposed personalized privacy preservation model and anonymous algorithm,this paper designs and implements a system to support the desensitization of medical image heterogeneous data.First of all,this paper analyzes the system requirements and describes the core functions and performance indicators of the system;then carries on the outline design of the system,defines the workflow,interface and key data structure of the system;then designs each subsystem and its module in detail and gives the concrete implementation.Finally,through the system test,this paper verifies that the medical image desensitization system meets the expected design.