Adapting Medical Image Processing Tasks to a Scalable Scientific Workflow System

摘要

In this paper, we present a web-based medical image processing scientific workflow system called DATAVIEW. This platform is implemented to satisfy the need of physicians to process their data by computer scientists and therefore, overcome the problem of delay for communicating between these two groups and speed up the processing which is very important in emergency conditions. For employing this workflow system, clinicians do not require to install any software tools and can create, save, share, reuse, and run their workflows only using a web browser without knowing the implementation details of the service. For employing the DATAVIEW workflow system for the medical image processing purpose, we extend this workflow system by integrating medical image software tools to this platform which is one of the challenges of this research. Also, in order to provide processing the data in parallel, we integrate high-end computing resource like grids to the system to speed up the processing, which is another challenge for this work. This is very important in medical imaging field because some software tools are very time consuming and in the urgent time, the physicians need to process multiple patients in parallel and get the results fast. As a case study, we choose epilepsy, which is one of the most common brain disorders.