Privacy-Preserving Patient-Centric Clinical Decision Support System on SVM

Abstract

A safe decision support estimation in health care system preserves the privacy of the patient data, the
decision estimation and the server side clinical support system. Clinical decision support system, which uses data
mining technique to help clinician make proper decisions, has received considerable attention nowadays. The advantages
of clinical decision support system include not only improving diagnosis accuracy it also reducing diagnosis time. With
rapid growth of clinical data generated day by day, the classification techniques can be utilized to hollow out valuable
information to improve a clinical decision support system. In this study, proposed a new privacy-preserving patient-centric
clinical decision support system, which helps clinician complementary to diagnose the risk of patients’ disease in a
confidentiality way. The past patients’ treatment history of the diseases are kept in the cloud environment which can be
used to prepare the SVM (Support Vector Machine) classifier without disclose any human being health data, and then the
trained classifier can be applied to compute the disease risk for new coming patients and also allow these patients to
retrieve the top-k disease names according to their own preferences. In addition to that, to protect the privacy of past
patients’ treatment data, a classical cryptographic methodology called TDES (Triple Data Encryption Standard) scheme is
proposed. The accuracy of the proposed system is evaluated via widespread simulations and also it demonstrates efficiency
of patient’s disease risk classification.