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.

医疗保健数据的安全性和隐私是医疗保健行业的重要方面，因为准确的诊断依赖于医疗专业人员访问患者医疗保健数据。同样，患者通常需要访问他们的数据。然而，确保敏感健康数据安全共享，同时优先考虑隐私至关重要。本文提出了一种称为基于四元数的神经网络的创新解决方案，即医疗保健环境中的高级数据安全架构（ADSAH），它将椭圆曲线密码学（ECC）与区块链机制和基于深度模糊的神经网络（DFBNN）相结合，以保护云安全存储的健康数据。所提出的方法首先使用编码器对输入医疗数据进行编码，然后使用 ECC 技术对编码数据进行加密。用于加密数据的密钥安全地存储在区块链框架中。密钥被分成块以增强安全性，并采用SHA算法来识别这些块内的密钥事件。这些关键事件随后存储在云存储系统中。利用改进的遗传算法来生成加密和解密密钥。该算法是专门为保护医疗保健数据而定制的。授权的患者或医生可以使用密钥来访问医疗数据，以解密和检索必要的信息。通过考虑时间和成本等因素来评估所提出的网络的性能，并与现有研究进行比较。评估显示了显着的改进，包括减少了加密和解密过程所需的时间，与之前的研究相比，交易和执行成本也有所降低。通过将 ECC 与区块链机制和 DNN 相结合，ADSAH 方法提供了一种先进的解决方案，可确保云存储的健康数据的安全性和隐私性。它提供强大的加密功能，有助于高效且经济高效地访问授权个人，同时保护敏感的健康信息。