医学影像大数据的存储与挖掘技术研究

Study on Big Medical Imaging Data Storage and Mining Techniques

数字化的医学影像通常使用医疗数字影像传输协议(DICOM),并通过医学影像储存与传输系统(PACS)来做储存、传输与管理。随着医学影像愈来愈多,如何有效率地处理海量医学影像数据变成一个问题。与一般的图像大数据处理技术相比,医学影像大数据所使用的标准和环境有所区别,值得注意。探讨了国内外医学影像大数据的存储和挖掘主要技术手段,并结合本实验室视网膜图像大数据存储与挖掘初步开展的工作,为医学影像大数据开发人员和管理人员提供借鉴。

Generally,digitalized medical images are stored,transmit and managed in picture archiving and communication systems（PACS） in compliance of digital imaging and communication in medicine（DICOM）. With the increasing of medical images,how to deal with the large amount of medical imaging data effectively becomes a problem. Compared with general image data processing technology,the utilization of medical imaging relies on some data standards and environment,which is worthy of attention. This paper aims to explore the storage and mining techniques of medical imaging mass data in combination with our preliminary experimental work in retinal image data storage and mining,thus providing reference for other researchers.