Improving the Execution Performance of FreeSurfer
- PDF / 571,327 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 89 Downloads / 235 Views
ORIGINAL ARTICLE
Improving the Execution Performance of FreeSurfer A New Scheduled Pipeline Scheme for Optimizing the Use of CPU and GPU Resources J. Delgado & J. C. Moure & Y. Vives-Gilabert & M. Delfino & A. Espinosa & B. Gómez-Ansón
# Springer Science+Business Media New York 2014
Abstract A scheme to significantly speed up the processing of MRI with FreeSurfer (FS) is presented. The scheme is aimed at maximizing the productivity (number of subjects processed per unit time) for the use case of research projects with datasets involving many acquisitions. The scheme combines the already existing GPU-accelerated version of the FS workflow with a task-level parallel scheme supervised by a resource scheduler. This allows for an optimum utilization of the computational power of a given hardware platform while avoiding problems with shortages of platform resources. The scheme can be executed on a wide variety of platforms, as its implementation only involves the script that orchestrates the execution of the workflow components and the FS code itself requires no modifications. The
scheme has been implemented and tested on a commodity platform within the reach of most research groups (a personal computer with four cores and an NVIDIA GeForce 480 GTX graphics card). Using the scheduled task-level parallel scheme, a productivity above 0.6 subjects per hour is achieved on the test platform, corresponding to a speedup of over six times compared to the default CPU-only serial FS workflow.
J. Delgado (*) Asociación para la Innovación en Análisis, Gestión y Procesamiento de Datos Científicos y Tecnológicos (INNDACYT), Avda. Europa, 20, Planta Baja Puerta D, 08907 Hospitalet de Llobregat (L’), Barcelona, Spain e-mail: [email protected]
FreeSurfer (FS) is a set of software tools for analysis and visualization of structural and functional brain imaging data developed at the Athinoula A. Martinos Center for Biomedical Imaging. FS is documented and freely available to download online http://surfer.nmr.mgh.harvard.edu/fswiki. The FS workflow for cortical reconstruction and volumetric segmentation of Magnetic Resonance Images (MRI) is a very commonly used tool in medical research of the brain. We will refer to this workflow as FS-MRI. It comprises two streams: the surface-based and the volume-based stream. The surfacebased stream calculates, for each point on the cortex, the cortical thickness, volume, surface area, and curvature (Dale et al. 1999; Fischl et al. 1999). The volume-based stream quantifies the volumes of the subcortical structures (Fischl et al. 2002, 2004). FS-MRI is structured as a pipelined stream of computation processes, whose execution is controlled by the recon-all shell script. The execution time of the workflow for a single MRI acquisition ranges from 10 to 30 h, depending on the hardware platform. Shortening this execution time is important for research projects which involve many MRI acquisitions or many subjects, as well as for potential future clinical applications.
J. Delgado : J. C. Moure : A. Espinosa
Data Loading...
