Case Western Reserve University researchers turned to GPUs running Jacket to develop a fast and robust Iterative Decomposition of water and fat with an Echo Asymmetry and Least-squares (IDEAL) reconstruction algorithm. The complete article can be found here.
The authors report that “GPU usage is critical for the future of high resolution, small animal and human imaging” and Jacket “enables GPU computations in MATLAB.”
Their research was performed on a desktop system with 32GB RAM, dual Intel Xeon X5450 3.0 GHz processors, an NVIDIA Quadro FX5800 (4GB RAM, 240 cores, 400 MHz clock), and MATLAB R2009a 64bit. Jacket v1.1, an older version, was used to produce these results.
Reconstruction tests with different sized images were performed to evaluate computation times for GPU and CPU implementations. Execution time was recorded for eleven repetitions, the first repetition was discarded and the remaining ten were averaged.
Here are a few of the images:
They found that GPUs running Jacket were faster than the CPU implementation for all image sizes and the advantage of GPUs improved with larger image sizes. GPUs performed 10X faster than the CPU implementation for 512 x 256 images.
The trajectory curves in the graph suggest the speedup provided by GPUs and Jacket improves as image sizes increase.
Before the authors utilized Jacket running on GPUs, reconstruction time exceeded acquisition time (1 hr. vs. 15 min.). After implementation, they were able to analyze large amounts of data in under 5 min. There data implies execution time for ultra-high resolution 2048×2048 data sets will take 47.8 min on the CPU and only 4.1 min on the GPU.
The Big Takeaway: If you need fast image processing, check out Jacket and GPUs. They can save you lots of processing time so that you can get back to doing your great work.
We want to thank David H. Johnson, Sreenath Narayan, Chris A. Flask, and David L. Wilson. They are doing outstanding work!
Comments 1
testing comments