| Abstrakt: |
Developing a secured and accurate disease diagnosis framework in the healthcare cloud systems are still remains one of the crucial problems in recent times. Due to the rapid growth of information and technology, it is highly essential to protect the patient health information against the unauthenticated users for ensuring the privacy and security. For this purpose, the different types of security approaches are developed in the conventional works, which are mainly focused on increasing the privacy of medical data stored in the cloud systems. However, it lacks with the major issues of increased computational overhead, communication cost, lack of security, complex mathematical modeling, and increased time consumption. Therefore, the proposed work objects to implement an intelligent and advanced privacy preserving framework, named as, lightweight signing signature based secured jellyfish data aggregation (LS3JDA) for ensuring the privacy of medical data in the healthcare cloud systems. The main contribution of this research work is to develop a new and lightweight privacy preservation model by incorporating the functions of both AI and signing signature algorithms for assuring data security in cloud systems. Moreover, it simplified the process of entire privacy preservation system with low computational burden and high data security. It also objects to accurately predict the type of disease based on the patients' medical history by using an advanced random forest (RF) machine learning methodology. The novel contributions of this work are, a message signing signature generation algorithm is used to strengthen the security of patients' medical data, and a jelly fish optimization (JFO) methodology is used to improve the process of data aggregation. The primary advantages of the proposed system are reduced processing time, low computational burden, and simple to deploy. For validating the results of the proposed model, several parameters include level of security, time, throughput, latency, signature cost, and communication overhead are assessed during evaluation. Moreover, the results are contrasted with some of the recent privacy preservation models for assuring the superiority of the proposed framework. Here, the overall processing time is reduced up to 1.5 ms, and communication overhead is reduced up to 100 bytes with the use of optimization integrated data aggregation model. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink