基于云计算的医疗大数据临床决策问题研究

Healthcare big data has become an important research direction in the medical field, and clinical decision-making is its important medical application. This project will focus on the research on clinical decision problem of healthcare big data based on cloud computing. Firstly, this project will study the abstraction and scheduling model of computing, storage, networking resources in medical field to improve resource utilization and management, which will lay a resource foundation for follow-up study of clinical decision-making problems. Secondly, to achieve massive clinical data classification, the project will study the clinical data classification model based on decision tree, which classifies clinical data according to its characteristics. Besides, the classified data will be distributedly stored on the big data platform, which provides the data basis for subsequent study of clinical decision-making problems. Thirdly, because of the single or multiple features that the clinical data have, clinical diagnosis decision algorithm based on features will be studied, which will become a foundation for clinical treatment decision research. Fourthly, to provide a reliable and effective treatment program for a patient, the project will study the clinical treatment decision algorithm based on the expectation value decision. Finally, the project will establish the software simulation platform based on C++ language and hardware simulation platform based on Openstack, where all kinds of models and clinical decision algorithms proposed will be simulated and validated, and the results will give guidance and feedback about model and algorithm design.

医疗大数据已经成为医疗领域的重要研究方向之一，而临床决策是医疗大数据的重要应用领域。本课题将研究基于云计算的医疗大数据临床决策问题。首先，本课题将研究医疗计算、存储、网络资源抽象与调度模型，提高医疗资源利用率与管控水平，为后续临床决策问题的研究提供资源层面的基础；其次，为实现对海量临床数据的分类，研究基于决策树算法的临床数据分类模型，按照数据特征对临床数据进行分类，并在大数据平台分布式存储分类后的临床数据，为后续临床决策问题的研究提供数据层面的基础；接下来，针对临床数据具有单一特征或多特征的特点，研究基于特征的临床诊断决策算法，作为临床治疗决策研究的基础。为制定可靠、有效的治疗方案，研究基于期望值决策的临床治疗决策算法；最后，建立基于C++语言的软件仿真平台和基于Openstack的硬件仿真平台，对本课题提出的各种模型和临床决策算法进行仿真与验证，并反馈指导模型和算法的设计。

医疗大数据已经成为医疗领域的重要研究方向之一，而临床决策是医疗大数据的重要应用领域。本课题将研究基于云计算的医疗大数据临床决策问题。首先，本课题将研究医疗计算、存储、网络资源抽象与调度模型，提高医疗资源利用率与管控水平，为后续临床决策问题的研究提供资源层面的基础；其次，为实现对海量临床数据的分类，研究基于决策树算法的临床数据分类模型，按照数据特征对临床数据进行分类，并在大数据平台分布式存储分类后的临床数据，为后续临床决策问题的研究提供数据层面的基础；接下来，针对临床数据具有单一特征或多特征的特点，研究基于特征的临床诊断决策算法，作为临床治疗决策研究的基础。为制定可靠、有效的治疗方案，研究基于期望值决策的临床治疗决策算法；最后，建立基于C++语言的软件仿真平台和基于Openstack的硬件仿真平台，对本课题提出的各种模型和临床决策算法进行仿真与验证，并反馈指导模型和算法的设计。课题相关成果可以应用到网络通信与人工智能领域，提高互联网多媒体内容传输效率和网络资源利用率。