Using GPGPU in MR Data Processing

0506976 - ÚPT 2020 RIV CZ eng C - Konferenční příspěvek (zahraniční konf.)
Mikulka, J. - Slíž, J. - Bartušek, Karel
Using GPGPU in MR Data Processing.
International Interdisciplinary PhD Workshop. Proceedings. Brno: VUT Brno, 2016, s. 102-106. ISBN 978-80-214-5387-6.
[International Interdisciplinary PhD Workshop. Brno (CZ), 12.09.2016-15.09.2016]
Institucionální podpora: RVO:68081731
Klíčová slova: GPGPU * data processing
Obor OECD: Radiology, nuclear medicine and medical imaging

The main tissue parameters targeted by MR tomography include, among others, relaxation times T-1 and T-2. This paper focuses on the computation of the relaxation time T-2 measured with the Spin Echo method. The maxima of measured echoes must be interleaved with an exponential function to compute the T2 relaxation. As this procedure needs to be repeated for each pixel of the scanned tissue, the processing of large images then becomes very intensive. This paper introduces the results provided by accelerated computation based on parallelization and carried out with a graphics card. By using the simple method of linear regression, we obtain a processing time of less than 36 ms. In the case of the Levenberg-Marquardt algorithm, the reconstruction was done in 96 ms. This period is at least 900 times shorter than that achievable with professional software.
Trvalý link: http://hdl.handle.net/11104/0298309