A Privacy-Preserving Intelligent Medical Diagnosis System Based on Oblivious Keyword Search

One of the concerns people have is how to get the diagnosis online without privacy being jeopardized. In this paper, we propose a privacy-preserving intelligent medical diagnosis system (IMDS), which can efficiently solve the problem. In IMDS, users submit their health examination parameters to the server in a protected form; this submitting process is based on Paillier cryptosystem and will not reveal any information about their data. And then the server retrieves the most likely disease (or multiple diseases) from the database and returns it to the users. In the above search process, we use the oblivious keyword search (OKS) as a basic framework, which makes the server maintain the computational ability but cannot learn any personal information over the data of users. Besides, this paper also provides a preprocessing method for data stored in the server, to make our protocol more efficient.

[1]  Bijan Raahemi,et al.  Importance of data mining in healthcare: A survey , 2015, 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM).

[2]  Pascal Paillier,et al.  Public-Key Cryptosystems Based on Composite Degree Residuosity Classes , 1999, EUROCRYPT.

[3]  Nitesh V. Chawla,et al.  Reliable medical recommendation systems with patient privacy , 2010, IHI 2010.

[4]  Julien Bringer,et al.  Biometric Identification over Encrypted Data Made Feasible , 2009, ICISS.

[5]  Elisa Bertino,et al.  State-of-the-art in privacy preserving data mining , 2004, SGMD.

[6]  Min Chen,et al.  iDoctor: Personalized and professionalized medical recommendations based on hybrid matrix factorization , 2017, Future Gener. Comput. Syst..

[7]  Insu Song,et al.  Anonymous Indexing of Health Conditions for a Similarity Measure , 2012, IEEE Transactions on Information Technology in Biomedicine.

[8]  Pim Tuyls,et al.  Efficient Binary Conversion for Paillier Encrypted Values , 2006, EUROCRYPT.

[9]  Rafail Ostrovsky,et al.  Secure two-party k-means clustering , 2007, CCS '07.

[10]  Chunxiao Jiang,et al.  Information Security in Big Data: Privacy and Data Mining , 2014, IEEE Access.

[11]  Stefan Katzenbeisser,et al.  Privacy-Preserving Recommendation Systems for Consumer Healthcare Services , 2008, 2008 Third International Conference on Availability, Reliability and Security.