Design And Implementation Of Zero- Footprint Medical Imaging Cloud Platform System

In the past decade,medical imaging visualization technology has developed rapidly,and a variety of new devices and technologies have emerged.At the same time,people are increasingly demanding reference to medical images.In particular,the increasing demand for convenient reading on mobile devices is high.With the advantages that cloud computing gradually shows in the medical field,the traditional medical image visualization system can only achieve the viewing operation of 2D medical images,and the image rendering processing operations are completed at the front end,and a large amount of image rendering processing will occupy a large number of front-ends.Memory affects the responsiveness and user experience of the request.Therefore,a new medical image review method is urgently required,namely,a zero-footprint medical image interpretation method.The image rendering processing is performed on the back-end processing,and the front-end is only responsible for sending requests and Display request results,etc.,to avoi d problems in the traditional medical image visualization system.This paper studies the design and implementation of zero-footprint medical imaging cloud platform system.The main research content is divided into three parts.The first is the data display module,which includes the design and implementation of front-end styles and page layouts,as well as the design and implementation of Java Script components such as file upload,image display,image rotation,image dragging and scaling,and then communication interaction modules.The data exchange between the front and back ends,and the study of the load balancing scheme under high concurrency scenarios in the cloud computing environment;the last is the data reading module,including the storage and retrieval of medical image files in the HDFS-based medical image storage system.The study of optimization programs.In this system,the data display module is mainly implemented by using HTML5,CSS,and Java Script.The implementation of the front-end style and page layout is based on the bootstrap style and the system requirements.The components of Java Script are used as the core part,and the back-end is invoked through the proxy.WCF services,etc.,perform communication and data exchange between the front and back ends of the WCF service application,and propose a dynamic load balancing scheme through research on Nginx;a data reading module uses the Sequence File technology to complete medical image file merging,and then merges the data.The file uses Map Reduce to index.Finally,through the functional test of the system,the main functions of the Web client of the system are tested,the test results are normal,and the basic requirements are met.Through non-functional test of the system,first compare the three load balancing algorithms that are respectively called by Nginx and the dynamic load balancing algorithm proposed by this paper.The test results show that the proposed dynamic load balancing algorithm is highly concurrency.The response efficiency is significantly higher than that of Nginx's own algorithm.Finally,the read and write time of the medical image file before and after the optimization of the Hadoop HDFS system,the name node memory consumption,and the data node memory consumption are displayed.The optimized system has a significantly faster response to reading and writing files,and the name node memory consumption and data node memory consumption are significantly reduced.