Medical big data access control model based on UPHFPR and evolutionary game

This paper discusses how to improve the accuracy of doctors' diagnosis and how to protect the security of patients' information. First, UPHFPR (Uncertain Probability Hesitant Fuzzy Preference Relationship) is applied to select more accurate target for doctors. The framework involved an information entropy to quantify the access request risks and privacy risks when doctors access clinical data. Based on the bounded rationality hypothesis, we build a multi-player evolutionary game model of risk access control, and analyze the participants' dynamic selection strategy and evolutionary stability. The simulation experiments suggest that UPPHFPR can help doctors choose the correct work objectives by integrating doctors' diagnostic opinions; we also incorporate the risk of doctor's access behavior into the evolutionary game's profit function, which can realize risk-adaptive access control. This model avoids the disclosure of clinical data and effectively protects the patients’ privacy.