Cloud Computing Assisted Mobile Healthcare Systems Using Distributed Data Analytic Model

作     者：Dhote, Sunita Baskar, S. Shakeel, P. Mohamed Dhote, Tejas 

作者机构：Department of Management Technology Shri Ramdeobaba College of Engineering and Management Nagpur Maharashtra India Department of Electronics and Communication Karpagam Academy of Higher Education Coimbatore Tamil Nadu India Faculty of Information and Communication Technology Universiti Teknikal Malaysia Melaka Melaka Malaysia Department of Mechanical Engineering - Engineering Mechanics Michigan Technological University Houghton MI USA 

出 版 物：《IEEE Transactions on Big Data》 (IEEE Trans. Big Data)

年 卷 期：2023年

页      面：1-12页

核心收录：

学科分类：0831[工学-生物医学工程（可授工学、理学、医学学位）] 1007[医学-药学(可授医学、理学学位)] 1004[医学-公共卫生与预防医学(可授医学、理学学位)] 100706[医学-药理学] 1002[医学-临床医学] 0808[工学-电气工程] 1001[医学-基础医学(可授医学、理学学位)] 08[工学] 0835[工学-软件工程] 0803[工学-光学工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

主　　题：Big data 

摘      要：Distributed cloud technologies enable mobile healthcare applications to support end users constantly. Data from electronic health records is made available through combination of user needs and heuristic mining. Because of inefficient data storage and reorganization, service and recommendation failures occur prematurely. A Distributed Data Analytics and Organization Model (DDAOM) is established in this article as a solution to the inefficiency of managing large amounts of data. Using this method, errors caused by performing several computations or storing large amounts of data in the medical field are minimized. Data organization and mining under predetermined schedules or factors provide information relevant to user services. One-to-many computations with varying input and output data allocations (for services) are executed in federated learning. The local input from several edges may be handled by allocating storage in a decentralized manner. The federated learning system uses the memory of past states to direct the allocation. Differentiating the states is necessary to allocate services and prevent mining in certain areas. With the help of realistic learning iterations, state management is maintained, guaranteeing the smooth deployment of services. Delays in storage and mining, uneven service provisioning, and service backlogs are used to evaluate the effectiveness of the suggested model. IEEE