Research On Privacy Protection Methods In Intelligent Medical Diagnosis System

As the technology of big data are gradually applied to the medical field,more and more people tend to get medical services online.However,what users are most concerned about is how to ensure that their health data will not be misused or stolen.In addition,the organizations that provide medical services are usually composed of multiple organizations that do not trust each other.Therefore,data security of data providers is also a key issue.Data providers have outsourced data on cloud platforms.This has become an irreversible trend.This paper proposes a privacy-preserving single-user intelligent diagnosis system that solves the problem of how to protect the data security of users and data providers in the context of single user and single data provider.This solution has been published in the SCI journal.Secondly,on the basis of a single-user system,we expand it to a multiple users and multiple data sources system,and propose a privacy-preserving multi-user intelligent medical diagnosis system,which can solve the problem of protecting the data security of users and multiple data providers.First of all,for single-user intelligent medical diagnosis system,this paper designs three phases based on Paillier encryption algorithm and oblivious keyword search,corresponding to three security algorithms.In the submit phase,the system generates the required secret key,and the user encrypts his own data and submits it to the server.During the preprocessing phase,the server encrypts the data in the database and generates an index.In the search phase,the server retrieves the user's request in the database to obtain the search result,that is,the diagnosis result,and sends it to the user.Then,based on the single-user system,this paper proposes a multi-user intelligent medical diagnosis system.The multi-user system also contains three phases,which are the setup phase,the preprocessing phase,and the search phase.It is mainly based on the Paillier encryption algorithm and the symmetrical searchable encryption technology.During the setup phase,the data provider generates keys for the user and the cloud platform respectively,and distributes them to them.During the preprocessing phase,the data provider indexes the database and sends the index to the cloud platform.In the search phase,the cloud platform first verifies whether the user is a authenticated user,and if it is a authenticated user,the cloud platform retrieves the diagnostic result for it and returns it to the user.Through theoretical proofs and experimental verification,the single-user and multi-user intelligent medical diagnosis system proposed in this paper can provide users with effective,reliable and safe medical diagnosis services.