Efficient access requests management for healthcare data with security and privacy-preserving

Data security and privacy-preserving are vital parts in the healthcare context for protecting confidential patient information, medical data, and history with local hospitals. Owing to single data type-based security, lack of policy-hiding strategies, and insider threats, conventional data security technology failed to guarantee data security. Thus, this work proposes a secure healthcare framework for ensuring the security of patient's healthcare data, realizing Access Control (AC) with the help of the data owner, and supporting normal and emergency scenarios. Primarily, to maintain secure data sharing, the patients are provided with a sub-secret during patient and doctor registration. After that, by utilizing the Recurrent and Convolution Layered Neural Network (RCLNN), the healthcare data is collected and categorized into two data types; then, the categorized data are encrypted using the Twisted Hessian Curve Cryptography (THCC) algorithm and Inversion Method (IM)-based embedding technique. Moreover, to provide authorization, an access policy is enforced. Hence, by utilizing the Standardized Padding Scheme-based Secure Hashing Algorithm-256 (SPS-SHA-256) algorithm, the AC policy is securely stored in the Blockchain (BC). Conversely, the user requests are classified using a Fuzzy Logic System (FLS) to provide controlled data access according to the risk evaluation and allow patients to define data access. Lastly, the outcomes displayed that for healthcare data security as well as privacy-preserving, the proposed AC model is highly secure and efficient.