الفهرس | Only 14 pages are availabe for public view |
Abstract As expectations for the data processing platforms to support processing for healthcare data, new services need to be deployed in healthcare services. Different needs from healthcare organizations raised the importance of the data to become reliable, high available, and high-performance processing. Additionally, the security and privacy of patient data have become a critical issue as a new need to keep health data secured on public clouds. Today, big data processing plays an active role in performing meaningful real-time analysis of the massive volume of health data to predict emergencies. Therefore, a first motivation is to go beyond the limitations of traditional data centres and provide a self-healing framework for healthcare data centres. A second possible motivation is to provide a framework for healthcare organizations to support real-time processing for massive healthcare data. The last possible motivation is enabling healthcare entities to use public cloud securely. This thesis presents a novel framework, for self-healing of healthcare data centres, and improving performance of healthcare data processing. The performance of the framework is improved by using a multi-agent architecture based on a distributed algorithm for data mining. Furthermore, the self-healing healthcare data centre is improved by taking into account reliability and availability during task execution jobs. Moreover, this thesis presents an intelligent algorithm called IFHDS to secure health data on a public cloud. IFHDS splits sensitive data into multiple parts according to sensitivity level, where each part is stored separately over distributed cloud storage. Finally, integrating all components in a novel framework to solve one of the most critical challenges in healthcare era, which is related to patient re-admission. |