Fast CT Reconstruction - Practical Performance via Parallelization

Parallel computing has been used to solve large-scale problems in many fields. While CT is being developed towards high-resolution, volumetric, dynamic and spectral imaging, datasets become increasingly large, and reconstruction speeds are often too slow. To meet this challenge, in 2004 Drs. Wang and Ni co-found a High performance Computing Lab, and have been working in this area ever since. In 2006, we designed and implemented the first parallel Katsevich algorithm. We have also parallelized EM, OS-EM, SART and OS-SART algorithms, respectively.