high performance computing on graphics processing units: hgpu.org

Fast image reconstruction is a critical requirement for an imaging modality to be adopted in the field of clinical and pre-clinical sciences. While programs become faster due to more powerful hardware, at the same time data size increases and the need for advanced—and often computational more demanding—reconstruction algorithms arises. A cheap way to achieve a major speed-up is to utilize modern graphics hardware capable of executing algorithms in a massively parallel manner. In this article, the open source library AGILE designed for image reconstruction in biomedical sciences is introduced. Its modular, object-oriented and templated design eases the integration of the library into user code. Furthermore, applications from the field of magnetic resonance imaging and fluorescence tomography are presented. As demonstrated, a speed-up of factor 10–30 is achievable with commodity graphics hardware for different reconstruction tasks.