SPL ENGINEERING CORE (FORMERLY THE BIRN SUBPROJECT)

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Organizational Note- The major event of this reporting period is the successful reorganization of BIRN (Biomedical Informatics Research Network) involving the separation of its principal cores into autonomous and independent research efforts. At the time of the last competitive renewal, Morphometry BIRN (mBIRN), Functional BIRN (fBIRN), and the NAC Engineering Core functioned as a single entity. Although the BIRN Core was well reviewed by the study section (score 130) and funded for the 5-year cycle, as part of an overall restructuring, NCRR requested a multi-year reorganization. Completed in 2005, the reorganization of mBIRN and fBIRN into distinct organizational entities resulted in the successful submission, review, and funding of new U24 consortium grants for mBIRN and fBIRN, respectively. The mBIRN consortium, led by Dr. Bruce Rosen at MGH, commenced independent operation on October 1, 2004 and the fBIRN consortium, led by Dr. Steven Potkin of UC Irvine, assumed operation on December 1, 2005. The partial-year efforts of the fBIRN are presented in the NAC Research Progress report and are also covered in the fBIRN progress report. The mBIRN project has an independent progress report and is not directly addressed here. This Core of the NAC effort, formerly known as the BIRN Core, will henceforth be known as the NAC Engineering Core. NAC Engineering Efforts- The NAC Engineering efforts center around the creation and maintenance of the software environment needed to support the research activities of NAC researchers, collaborators, and users. Much of this activity is embodied in the 3D Slicer application software. The Slicer software provides a fully open source, cross platform, "point and click" end user application software environment for neuroimage analysis and related imaging applications. Within the NAC effort, Slicer is a critical medium for collaboration; a common usage pattern is required for novel algorithms to be implemented as Slicer Modules and delivered to clinical and applied science collaborators for use ''with their existing data and for the same application.'' This transitional pattern is possible as a consequence of ongoing efforts to ensure that algorithm developers are educated and supported in their use of the Center's common software practices. Another important aspect of this transitional pattern is the shared computational infrastructure which includes high performance graphics workstations, multi-CPU compute servers, tera-scale disk storage, and gigabit networking. This shared infrastructure permits algorithm developers to have controlled access to clinical and research data for developing and testing new approaches. The new software thus developed is easily tested by end-users who are running inside the same software environment. This rapid exchange between developers and users is crucial to effective development. During this reporting period, the NAC was grateful to accept an equipment supplement that permitted the purchase and installation of 70 new high-end workstations. This much-needed upgrade allowed the Center to migrate from the Sun Solaris operating system to a 64-bit Linux environment. As with all major upgrades, significant effort was dedicated to ensuring a graceful transition, particularly with respect to Slicer and its supporting software packages. We are pleased to report that this transition has been effected with minimal disruption, in large part owing to the excellent engineering represented in the software packages on which Slicer relies (VTK, ITK, Tcl/Tk). The further addition of a set of 16-CPU compute servers and a 200 node compute cluster are expected to result in additional demands for software maintenance and porting to enable further advanced research. An enduring feature of the NAC Engineering effort is an ongoing and rich collaboration with national-level consortia. In particular, the mBIRN and fBIRN efforts serve to define and develop best practices for image acquisition, calibration, data-sharing, and database/informatics; the NAC Engineering Core remains active in these research efforts with technology and benefits flowing in both directions. The National Alliance for Medical Image Computing, NA-MIC, has a similar synergistic relationship in the areas of algorithm development and software engineering methodologies. NA-MIC provides an important mechanism for direct interaction with the architects, developers, and maintainors of the Insight Software Toolkit (ITK), which in recent years has become the platform for much of the algorithm development work in NAC.