Output details
11 - Computer Science and Informatics
University College London
Cartesian SENSE and k-t SENSE reconstruction using commodity graphics hardware
<28> Introduces a new concept for interacting in real-time with algorithm parameters in magnetic resonance image reconstruction. Users can interact with parameters previously regarded as “fixed”, to determine the optimum values for a specific application. Speed achieved by physics-informed linear algebra algorithm and GPU implementation. One of three original papers forming the basis for the recent Open Source Gadgetron software for streaming reconstruction (gadgetron.sourceforge.net, 2000+ downloads, 46% in USA). Interfaced to Siemens clinical MRI scanners for cardiac and interventional research applications. Rigour: Algorithmic timing comparisons were presented along with a representative selection of images and data time-slices.