Disease Risk Prediction Using Data Mining with Privacy Preservation of Data

  • Rutvik Mahajan, Parnal Tambat, Darshan Shinde, Kaustubh Yewale, Pooja Vengurlekar


Data mining-driven disease risk prediction has become one of the important topics in the field of e-healthcare. With the widespread use of hospital information system, there is a huge amount of generated data which can be used to improve healthcare service. However, without the security and privacy assurances, disease risk prediction cannot continue to flourish. To address this challenge, an efficient and privacy-preserving disease risk prediction model for e-healthcare is proposed. Compared with the up-to-date works, the proposed work comprehensively achieves two phases of disease risk prediction - disease model training and disease prediction, while ensuring privacy preservation. NAIVE BAYES algorithm is introduced to compute the classification result. The model makes use of the ABE algorithm for encryption and also demonstrates prediction of disease with the AID data set using the SVM algorithm. Besides, extensive performance evaluations demonstrate that our proposed model attains outstanding efficiency advantage and hence is more suitable for real-time e-healthcare, especially medical emergency.