Parallel Computation in Medical Imaging Applications
Yasser M. Kadah, Khaled Z. Abd‐Elmoniem, Aly A. Farag · International Journal of Biomedical Imaging · 2011
There is currently a rapidly growing interest in parallel computation application in various medical imaging and image processing fields.This trend is expected to continue growing as more sophisticated and challenging medical imaging, image processing, and high-order data visualization problems are being addressed.The ongoing cost drop in computational tools and their wide accessibility play a center role as well.Given its short history, this area is still not a welldefined scientific discipline.The selected topics and papers for this special issue shed more light on various aspects of this expanding field and its potential in accelerating medical imaging applications.This special issue contains eleven papers covering various imaging modalities including MRI, CT, X-ray, US, and optical tomography.The papers demonstrated the potential of parallel computation in medical imaging and visualization in a wide range of applications including image reconstruction, image denoising, motion estimation, deformable registration, diffeomorphic mapping, and modeling.In the paper entitled "CUDA-accelerated geodesic raytracing for fiber tracking," E. van Aart et al. present an accelerated algorithm for brain fiber tracking.Noninvasive diffusion weighted imaging followed by reconstructing the brain fiber structure provides a unique way to inspect the complex structures inside the brain in a microscopic level.However, these processes are computationally expensive.The proposed algorithm utilizes the parallel structure of a graphics processing unit in combination with the CUDA platform to substantially accelerate the execution time of the fiber tracking by a factor up to 40 times compared to a multithreaded CPU implementation.