Implementing Ultrasound Beamforming on the GPU using CUDA

Lars Gronvold
Department of Engineering Cybernetics, Norwegian University of Science and Technology
Science And Technology (2008) Issue: May


   title={Implementing Ultrasound Beamforming on the GPU using CUDA},

   author={Gr{o}nvold, L.},



Download Download (PDF)   View View   Source Source   



Todays ultrasound equipment consists mainly of a PC that is attached to several large cards that process the received signals in hardware. These cards take up a lot of space and are costly to develop. As processing power in PC’s increase it is possible move some of this signal processing from specialized hardware to standard PC hardware. Such a transition from hardware over to software could open up new possibilities for new and more advanced signal processing that could increase the image quality. The task is mainly about focusing ultrasound beams(beamforming) and can be divided in a theoretical and a practical part. First, study current and “future” ultrasound beamforming technology. Secondly, design algorithms and implement them on the NVIDIA Tesla GPU card. This GPU card provides a PC with high performance parallel number crunching.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: