Suranarong Kamtasila MEng, Presenter: Nothing to Disclose

Krongrat Kangwanklai BS, MS, Abstract Co-Author: Nothing to Disclose

An Image Archiving System can be characterized as a Big Data architecture.It is considered based on the growth of Modality data, various data structure,a need for faster response time, a longer storage duration, and regulatory requirements to maintain personal data for a very long period of time.This research focuses on exploring a suitable data architecture that meets the ISMS ISO 27001 needs and is based on the principles of Big Data architecture and Cloud.

Based on the ISO 27001 requirements, the medical data, its DBMS, and log files are our most important concern.An archiving system must have no limitation on the storage.The data must be encrypted, must not be tampered with and is accessible only through the specified applications.The file system must be fault-tolerance. The backup facilities must be adequate and its copies must be available in both a main site and a remote location. The system must be able to recover from major failures and support a business continuity management. We use Hadoop framework to ensure that the above requirements are met.The main reasons that HDFS is selected are it can support the Big Data architecture, stream data access, large data set, and simple coherency model.    

There are three Clouds. Two of them are designed for storage and the third one is for applications. These Clouds will be resided in two locations and connected through VPN.  At each data center, it is comprised of two Hadoop framework, one cluster of MySQL, one cluster of EJBCA, and one cluster of application. There is DICOMSync at each location which acts as a connecting point. It will compress and send the data from one DICOMSync to another, decompress and forward the data to the storage cluster.  

For ISO 27001 requirements, it is required that ISMS data are secured.The business continuity management process is implemented. The data is encrypted to maintain the data confidentiality. The log files are kept systematically to enable effective monitoring and traceability. The application interfaces are readily available to support research activities. All these are designed and implemented using the open source software.

Kamtasila, S, Kangwanklai, K, ISO 27001 - Implementing Medical Imaging Archiving Using Big Data Architecture: Open Source Approach.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13017289.html