Reconstructing 3D parameter maps of huge volumes entirely on the GPU is highly desirable due to the offered computation speed-up. However, GPU memory restrictions limit the coverable volume. To overcome this limitation, a double-buffering strategy in combination with model-based parameter quantification and 3D-TGV regularization is proposed. This combination warrants whole volume reconstruction while maintaining the speed advantages of GPU-based computation. In contrast to sequential transfers, double-buffering splits the volume into blocks and overlaps memory transfer and kernel execution concurrently, hiding memory latency. The proposed method is able to reconstruct arbitrary large volumes within 5.3 min/slice, even on a single GPU.
Introduction
Scan time reduction of MR parameter quantification with model-based reconstruction1,2,3,4,5 usually involves iterative optimization techniques. Computationally this leads to long reconstruction times coupled with possibly huge memory requirements. Recently, GPU acceleration for model-based reconstruction from highly subsampled data acquisition was proposed3,4,5. These approaches suffer from memory limitations of the available GPUs and are thus applied slice-by-slice in 2D3,4 or on small volumes5. The aim of the present work is to overcome memory limitations by making use of double-buffering6 which enables reconstruction of arbitrarilly large volumes on even a GPU. Double-buffering was recently applied to the non-uniform FFT (NUFFT)7. Here, we extend this approach to accelerate computation of a complete iterative reconstruction problem, i.e. a previously developed iteratively regularized Gauss-Newton (IRGN) algorithm for 3D T1-mapping from accelerated golden-angle radial-stack-of-stars data using total-generalized-variation (TGV) constraints5. For evaluation purposes the reconstruction times and accordances in image quality of the proposed strategy are compared against the unoptimized reconstruction. The proposed algorithm is implemented in Python/PyOpenCL8 and is therefore platform independent.This work is funded and supported by the Austrian Science Fund (FWF) under grant “SFB F32‐N18” (SFB “Mathematical Optimization and Applications in Biomedical Sciences”); NVIDIA Corporation Hardware grant support; Oliver Maier is a Recipient of a DOC Fellowship (24966) of the Austrian Academy of Sciences at the Institute for Medical Engineering at TU Graz.
1. Block KT, Uecker M, Frahm J. Model-Based Iterative Reconstruction for Radial Fast Spin-Echo MRI. IEEE Traansactions on Medical Imaging, Vol. 28, No. 11, November 2009
2. Sumpf TJ, Uecker M, Boretius S, Frahm J. Model-based nonlinear inverse reconstruction for T2 mapping using highly undersampled spin-echo MRI. J Magn Reson Imag 2011; 34(2):420–428
3. Roeloffs, V. , Wang, X. , Sumpf, T. J., Untenberger, M. , Voit, D. and Frahm, J. (2016), Model‐based reconstruction for T1 mapping using single‐shot inversion‐recovery radial FLASH. Int. J. Imaging Syst. Technol., 26: 254-263. doi:10.1002/ima.22196
4. Wang, X. , Roeloffs, V. , Klosowski, J. , Tan, Z. , Voit, D. , Uecker, M. and Frahm, J. (2018), Model‐based T1 mapping with sparsity constraints using single‐shot inversion‐recovery radial FLASH. Magn. Reson. Med, 79: 730-740. doi:10.1002/mrm.26726
5. Maier O., Schoormans J., Schloegl M., et al. Rapid T1 quantification from high resolution 3D data with model‐based reconstruction. Magn Reson Med. 2018;00:1–18. DOI: 10.1002/mrm.27502
6. Harris M. How to Overlap Data Transfers in CUDA C/C++. https://devblogs.nvidia.com/how-overlap-data-transfers-cuda-cc/#disqus_thread. Website accessed on 31.10.2018
7. Smith DS, Sengupata S, Smith SA, Welch EB. Trajectory optimized NUFFT: Faster non‐Cartesian MRI reconstruction through prior knowledge and parallel architectures. Magn Reson Med 2018; DOI: 10.1002/mrm.27497
8. Klöckner A, Pinto N, Lee Y, Catanzaro B, Ivanov P, Fasih A. PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation. Parallel Computing 2012; 38: 157-174.
9. Knoll, F.; Schwarzl, A,; Diwoky, C.; Sodickson DK.: gpuNUFFT - An Open-Source GPU Library for 3D Gridding with Direct Matlab Interface. Proc ISMRM p4297 (2014).
10. Lesch A, Schloegl M, Holler M, et al. RUltrafast 3D Bloch‐Siegert B+1‐mapping using variational modeling. Magn Reson Med. 2008. doi:10.1002/mrm.27434.