Abstract
We present a CUDA implementation for Kepler and Maxwell GPU generations of neuroimaging registration based on the NiftyReg open-source library [1]. A wide number of strategies are deployed to accelerate the code, providing insightful guidelines to exploit the massive parallelism and memory hierarchy within emerging GPUs. Our efforts are analyzed from different perspectives: Acceleration, numerical accuracy, power consumption and energy efficiency, to identify potential scenarios where performance per watt can be optimal in large-scale biomedical applications. Experimental results suggest that parallelism and arithmetic intensity represent the most rewarding ways on the road to high performance bioinformatics when power is a major concern.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Modat, M.: NIFTYREG - a library to perform rigid, affine and non-linear registration of NIfTI images. http://sourceforge.net/projects/niftyreg/
NIfTI: the NIfTI format home page. http://nifti.nimh.nih.gov
Clay, R.: Functional magnetic resonance imaging: a new research tool (2007). www.apa.org/research/tools/fmri-adult.pdf
DICOMNIFTI: a tool for converting DICOM files into the NIfTI data format. http://cbi.nyu.edu/software/dinifti.php. Accessed October 2013
NIFTILIB: input/output libraries for NIfTI-1 neuroimaging data format. http://niftilib.sourceforge.net
CMIC: the NifTK software platform. http://www.niftk.org
Modat, M., Ridgway, G.R., Taylor, Z.A., Lehmann, M., Barnes, J., Hawkes, D.J., Fox, N.C., Ourselin, S.: Fast free-form deformation using graphics processing units. Comput. Methods Program. Biomed. 98(3), 278–284 (2010)
McNamee, J.: A comparison of methods for accurate summation. ACM SIGSAM Bull. 38, 1–7 (2004)
Arduino: an open-source electronics platform based on easy-to-use hardware and software. https://www.arduino.cc/en/Main/ArduinoBoardMega2560
Adafruit: INA219 current sensor breakout. https://learn.adafruit.com/adafruit-ina219-current-sensor-breakout
Ziegler, S., Woodward, R., Iu, H., Borle, L.: Current sensing techniques: a review. IEEE Sens. J. 9(4), 354–376 (2009)
Philips: I2C-bus specification and user manual. Philips Semiconductors (2014)
Igual, F., Jara, L., Gómez, J., Piñuel, L., Prieto, M.: A power measurement environment for PCIe accelerators. Comput. Sci. Res. Dev. 30, 115–124 (2015)
Acknowledgments
This work was supported by the Ministry of Education of Spain under Project TIN2013-42253-P and by the Junta de Andalucia under Project of Excellence P12-TIC-1741. We thank Javier Cabero and Pablo Sánchez for their work on preliminary versions of these CUDA implementations. We also thank Marc Modat from University College London, for his support when using the NiftyReg library. We also thank Nvidia for hardware donations within GPU Education Center 2011–2016 and GPU Research Center 2012–2016 awards.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Álvarez, F.N., Cabrera, J.A., Chico, J.F., Pérez, J., Ujaldón, M. (2016). Neuroimaging Registration on GPU: Energy-Aware Acceleration. In: Ortuño, F., Rojas, I. (eds) Bioinformatics and Biomedical Engineering. IWBBIO 2016. Lecture Notes in Computer Science(), vol 9656. Springer, Cham. https://doi.org/10.1007/978-3-319-31744-1_55
Download citation
DOI: https://doi.org/10.1007/978-3-319-31744-1_55
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-31743-4
Online ISBN: 978-3-319-31744-1
eBook Packages: Computer ScienceComputer Science (R0)