Correlation analysis on GPU systems using NVIDIA’s CUDA

• Verfasst von: Gembris, Daniel [VerfasserIn]
• Verfasst von: Neeb, Markus [VerfasserIn]
• Verfasst von: Gipp, Markus [VerfasserIn]
• Verfasst von: Kugel, Andreas [VerfasserIn]
• Verfasst von: Männer, Reinhard [VerfasserIn]
• Titel: Correlation analysis on GPU systems using NVIDIA’s CUDA
• Verf.angabe: Daniel Gembris, Markus Neeb, Markus Gipp, Andreas Kugel, Reinhard Männer
• Jahr: 2011
• Umfang: 6 S.
• Fussnoten: Gesehen am 24.10.2023 ; Online veröffentlicht: 17. Juni 2010
• Titel Quelle: Enthalten in: Journal of real-time image processing
• Ort Quelle: Berlin : Springer, 2006
• Jahr Quelle: 2011
• Band/Heft Quelle: 6(2011), 4 vom: Dez., Seite 275-280
• ISSN Quelle: 1861-8219
• DOI: doi:10.1007/s11554-010-0162-9
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: BOLD
• Sach-SW: Correlation and regression analysis
• Sach-SW: fMRI
• Sach-SW: FPGA
• Sach-SW: Graphics processing unit (GPU)
• Sach-SW: Time series analysis
• K10plus-PPN: 1867432498

Abstract

Functional magnetic resonance imaging allows non-invasive measurements of brain dynamics and has already been used for neurofeedback experiments, which relies on real time data processing. The limited computational resources that are typically available for this have hindered the use of connectivity analysis in this context. A basic, but already computationally demanding analysis method of neural connectivity is correlation analysis that computes all pairwise correlations coefficients between the measured time series. The parallel nature of the problem predestines it for an implementation on massive parallel architectures as realized by GPUs and FPGAs. We show what performance benefits can be achieved when compared with current desktop CPUs. The use of correlation analysis is not limited to brain research, but is also relevant in other fields of image processing, e.g. for the analysis of video streams.