Secure and privacy in healthcare data using quaternion based neural network and encoder-elliptic curve deep neural network with blockchain on the cloud environment

【Author】 Suganthi, P.; Kavitha, R.

【Source】SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES

【影响因子】1.214

【Abstract】The security and privacy of healthcare data are crucial aspects within the healthcare industry, as accurate diagnoses rely on medical professionals accessing patient healthcare data. Similarly, patients often require access to their data. However, ensuring that sensitive health data is shared securely while prioritizing privacy is essential. This paper proposes an innovative solution called the quaternion based neural network, Advanced Data Security Architecture in Healthcare Environment (ADSAH), which combines Elliptical curve cryptography (ECC) with a blockchain mechanism and a Deep Fuzzy Based Neural Network (DFBNN) to safeguard cloud-stored health data. The proposed approach begins by encoding the input medical data using an encoder and then encrypting the encoded data using ECC techniques. The secret key for encrypting the data is securely stored within a blockchain framework. The key is divided into blocks to enhance security, and the SHA algorithm is employed to identify key events within these blocks. These key events are subsequently stored in a cloud storage system. A modified genetic algorithm is utilized to generate the encryption and decryption key. This algorithm is explicitly tailored to secure healthcare data. Authorized patients or physicians can access medical data using the secret key to decrypt and retrieve the necessary information. The performance of the proposed network is evaluated by considering factors such as time and cost and is compared against existing studies. The evaluation demonstrates notable improvements, including a reduction in the time required for the encryption and decryption process, as well as a decrease in transaction and execution costs when compared to previous research. By incorporating ECC with a blockchain mechanism and DNN, the ADSAH approach offers an advanced solution for ensuring the security and privacy of cloud-stored health data. It provides robust encryption and facilitates efficient and cost-effective access to authorized individuals while safeguarding sensitive health information.

【Keywords】Healthcare data; security; deep neural network; improved quaternion based neural network; blockchain mechanism; elliptic curve cryptography

【发表时间】2023 23-Sep

【收录时间】2023-10-21

【文献类型】实验仿真

【主题类别】

区块链应用-实体经济-医疗领域