Grid-Based Interactive Decision Support in Biomedicine

A huge gap exists between what we know is possible with today's machines and what we have so far been able to finish. —Donald Knuth 1.1 INTRODUCTION The challenges discovered when studying humans as complex systems, from a biomedical viewpoint (from cells to interacting individuals), cover the whole spectrum from genome to health and cross temporal and spatial scales [1]. This includes studying biomedical issues using multiscale and multiscience models and techniques all the way from genomics to the macroscopic medical scale. This is also aggravated by the continuous increase in the amount of digital data produced by modern high-throughput biomedical detection and analysis systems. As reported by Hey et al., it is expected that larger amounts of digital data will be generated by next generations of large scale, collaborative e-Science experiments [2]. New experiments in science and engineering will cover the whole spectrum, from the simulation of complete biological systems, to cutting-edge research in bioinformatics. At the macroscopic scale, for instance, there are research efforts in biomedical informatics that are gradually pushing the boundaries of the state of the art, moving from monolitic software architectures to building more generic components. Such efforts normally leverage object-oriented and distributed component architectures to encapsulate or wrap legacy data in order to improve application interoperability and scalability [3, 4]. This allows for enhanced data and process flow at the macroscopic level, where models such as DICOM provide support for data acces from work stations to archiving and communications systems and back to hospitals' information systems.