GPU acceleration for digitally reconstructed radiographs using bindless texture objects and CUDA/OpenGL interoperability

This paper features an advanced implementation of the X-ray rendering algorithm that harnesses the giant computing power of the current commodity graphics processors to accelerate the generation of high resolution digitally reconstructed radiographs (DRRs). The presented pipeline exploits the latest features of NVIDIA Graphics Processing Unit (GPU) architectures, mainly bindless texture objects and dynamic parallelism. The rendering throughput is substantially improved by exploiting the interoperability mechanisms between CUDA and OpenGL. The benchmarks of our optimized rendering pipeline reflect its capability of generating DRRs with resolutions of 20482 and 40962 at interactive and semi interactive frame-rates using an NVIDIA GeForce 970 GTX device.

[1]  Jos B. T. M. Roerdink,et al.  X-ray volume rendering by hierarchical wavelet splatting , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[2]  Marwan Abdellah,et al.  High Performance GPU-Based Fourier Volume Rendering , 2015, Int. J. Biomed. Imaging.

[3]  Zhiyi Yang,et al.  Parallel Image Processing Based on CUDA , 2008, 2008 International Conference on Computer Science and Software Engineering.

[4]  Arvid Lundervold,et al.  Advances in medical imaging , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[5]  Marwan Abdellah,et al.  Offline large scale fourier volume rendering on low-end hardware , 2014, 2014 Cairo International Biomedical Engineering Conference (CIBEC).

[6]  Klaus Mueller,et al.  Overview of Volume Rendering , 2005, The Visualization Handbook.

[7]  Atef Z. Elsherbeni,et al.  Utilization of CUDA-OpenGL interoperability to display electromagnetic fields calculated by FDTD , 2011, CEM'11 Computational Electromagnetics International Workshop.

[8]  Charles Hansen,et al.  The Visualization Handbook , 2011 .

[9]  M. Abdellah,et al.  Constructing a functional Fourier volume rendering pipeline on heterogeneous platforms , 2012, 2012 Cairo International Biomedical Engineering Conference (CIBEC).

[10]  T. Todd Elvins,et al.  A survey of algorithms for volume visualization , 1992, COMG.

[11]  Daniel Rueckert,et al.  Fast generation of digitally reconstructed radiographs using attenuation fields with application to 2D-3D image registration , 2005, IEEE Transactions on Medical Imaging.

[12]  Yoel Shkolnisky,et al.  3D discrete X-ray transform , 2004 .

[13]  Nicholas Wilt,et al.  The CUDA Handbook: A Comprehensive Guide to GPU Programming , 2013 .