More IMPATIENT: A Gridding-Accelerated Toeplitz-based Strategy for Non-Cartesian High-Resolution 3D MRI on GPUs
| Autor: | Maojing Fu, Justin P. Haldar, Wen-mei W. Hwu, Xiao-Long Wu, Joseph L. Holtrop, Fan Lam, Bradley P. Sutton, Zhi-Pei Liang, Nady Obeid, Jiading Gai |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Computer Networks and Communications
Computer science Computation ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Parallel computing Iterative reconstruction Toeplitz matrix Article Theoretical Computer Science law.invention CUDA Artificial Intelligence Hardware and Architecture law Cartesian coordinate system Massively parallel Throughput (business) Software ComputingMethodologies_COMPUTERGRAPHICS |
| Zdroj: | Journal of parallel and distributed computing. 73(5) |
| ISSN: | 0743-7315 |
| Popis: | Several recent methods have been proposed to obtain significant speed-ups in MRI image reconstruction by leveraging the computational power of GPUs. Previously, we implemented a GPU-based image reconstruction technique called the Illinois Massively Parallel Acquisition Toolkit for Image reconstruction with ENhanced Throughput in MRI (IMPATIENT MRI) for reconstructing data collected along arbitrary 3D trajectories. In this paper, we improve IMPATIENT by removing computational bottlenecks by using a gridding approach to accelerate the computation of various data structures needed by the previous routine. Further, we enhance the routine with capabilities for off-resonance correction and multi-sensor parallel imaging reconstruction. Through implementation of optimized gridding into our iterative reconstruction scheme, speed-ups of more than a factor of 200 are provided in the improved GPU implementation compared to the previous accelerated GPU code. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect