With the rapid development of IT data driven experiments and simulations have become very advanced and complicated. The amount and complexity of scientific data increase exponentially. Nowadays the challenge is to efficiently support scientists who generate more data than they can possibly look at and understand. This requires not only high-performance visualization and feature extraction techniques but also efficient and intuitive management of available data and underlying computational resources. Consequently, Service Oriented Architectures and workflow management systems are becoming popular solutions for the deployment of e-Science Infrastructures aimed to assist in exploration and analysis of large scientific data. In this paper, we compare two transport models of workflow execution for data-intensive medical visualizations that rely on web services. The image-based analysis of vascular disorders served as a case study for this project. We applied a service oriented approach to construct distributed visualization pipelines, which allow visualization experts to develop visualizations to view and interact with large medical data sets. Moreover, end-users (i.e., medical specialists) can explore these visualizations irrespective of their geographical location and available computing resources. The paper reports on the current implementation status and presents our main findings.
Koulouzis, S., Zudilova-Seinstra, E., & Belloum, A. (2010). Data transport between visualization web services for medical image analysis. In Procedia Computer Science (Vol. 1, pp. 1727–1736). Elsevier B.V. https://doi.org/10.1016/j.procs.2010.04.194