Hybrid QRE-RHIBE Encryption with Capability Based Access Control Model and Optimized ENN for Secure EHR Retrieval and Disease Detection in a Healthcare Cloud

Healthcare practitioners use Electronic Health Records (EHRs), which are digital records of a patient’s medical history that are kept up to date, to manage patient care. Integration of cloud computing in healthcare enables efficient storage, retrieval, and analysis of EHRs while ensuring robust security and enabling advanced disease detection. These days, security and control over data access are some of the main issues with cloud storage, particularly in the medical industry. To avoid these challenges Optimized Deep learning approach with hybrid cryptography algorithm is developed to secure EHR retrieval and disease detection. Initially, EHR data is collected and pre-processed using Linear Non-Gaussian Acyclic Model (LiNGAM) to fill the missing values and Local Context Normalization (LCN) approach to normalize the data. Then Modified Elman Neural Network (MENN) is used to detect the disease. To enhance the performance of Elman Neural Network, Social Spider Optimization Algorithm (SSOA) is utilized for optimally selecting the weight decay and dropout rate. Following that, to improve the security of data and safeguard patient privacy in the cloud server, a QRE-RHIBE is utilized. Quantum Resistant Encryption is used to generate a keys and the RHIBE is used for both encryption and decryption process based on the generated key. Then for accessing the data stored in the cloud server, Capability based Access Control Mechanism (CBACM) is employed for identifying and removing unwanted users’ access. The model that employed, which achieves an accuracy of 98.53%, precision of 98.02% and execution time of 1.2 sec. This approach ensures data privacy, precise access control, and enhanced diagnostic accuracy which promoting trust and innovation in healthcare systems.