Fast parallel analysis of dynamic contrast-enhanced magnetic resonance imaging on GPGPU

Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) is adopted as an important diagnostic method to photograph organs in a body. However, its long analysis time prevents doctors from performing timely diagnosis and treatment. This paper proposes a parallel implementation of DCE-MRI analysis which exposes the parallelism of the fitting procedure. The implementation is mapped and executed on GPGPUs. When compared with the original sequential analysis procedure, we are able to speed up the analysis runtime by 125X.