Proc. SPIE. 8674, Medical Imaging 2013: Advanced PACS-based Imaging Informatics and Therapeutic Applications
KEYWORDS: Image compression, Visualization, Image processing, Parallel processing, Medical imaging, Java, Local area networks, Computer architecture, 3D image processing, Picture Archiving and Communication System
Large volumes of data are common in medical imaging and require special applications that enable fast and reliable processing. High performance computing enables such processing by running applications in parallel. This work presents a cluster architecture for the parallel processing of DICOM datasets, where the functionality is not specified by the application itself but rather by a set of plugins. Results show that processing times reduce with each added node while network latency is kept stable.