به اشتراک گذاری منابع داده بزرگ پزشکی
Abstract
I. Introduction
II. Related Theories and Methods
III. Parallel Optimization of Deep Learning Algorithms for Medical Big Data Processing
IV. Analysis of Regional Information Resource Sharing Mechanism Based on Medical Big Data
V. Medical Big Data Visualization Implementation
Authors
Figures
References
Abstract
With the development of information technology, the informationization of the medical industry is also constantly developing rapidly, and medical data is growing exponentially. In the context of ‘‘Big Data +’’, people began to study the application of data visualization to medical data. Data visualization can make full use of the human sensory vision system to guide users through data analysis and present information hidden behind the data in an intuitive and easy-to-use manner. This paper first introduces the workflow of DBN, a deep learning algorithm, and summarizes the computational characteristics of the algorithm. The classification function is translated into an assembler using an instruction set-based assembly language, and the program is evaluated for performance. Secondly, based on the Hadoop ecosystem, this paper analyzes the BDMISS system for big data medical information resource sharing. Based on the system’s requirements and functional positioning, from the medical information collection and sharing, data mining and knowledge management level, the big data medical service system is constructed. Based on the semantic network and ontology theory, big data mining technology and the design of ‘‘medical cloud’’, the resource sharing mechanism is analyzed. Based on the Spring MVC framework, using Echarts, HCharts and other data visualization technology, according to the design of specific modules, the visualization and display of medical data is realized, which has certain promotion effect on the research and development of medical big data visualization analysis.
Introduction
As the informationization of the medical industry continues to develop rapidly, medical data has grown exponentially, and medical big data has brought tremendous pressure on existing hospital information systems. With the emergence of various unstructured data, traditional medical information systems cannot meet the requirements of big data in terms of storage space, storage speed, storage structure, etc., some data is lost, resulting in loss of valuable medical data [1]–[۳]. The data integrity of the system is not enough, and the data processing speed is slow, which cannot meet the user’s demand for data visualization display [4]. In recent years, the information and communication industry of medical and health has ushered in the opportunity of development. The hospital information management system, public health service platform, telemedicine, mobile medical, and information equipment have formed a scale of 100 billion yuan. Medical informatization is no longer limited to transactional tasks such as hospital information management systems [5], [6]. Applications and research based onInternet of Things and cloud computing continue to deepen, and medical informatization begins to develop in the areas of process optimization and service innovation [7]. The Internet of Things and cloud computing have changed the patterns and paths of medical information services, optimized medical service processes, improved the efficiency of medical services, and transformed medical information sharing and service models [8], [9]. Deep learning not only improves people’s lives in traditional industries, but as people’s attention to physical health increases, machine learning algorithms closely related to deep learning have emerged in the health care field.