Clemens Alexander Maaß, Abstract Co-Author: Nothing to Disclose

Marc Kachelrieß PhD, Presenter: Nothing to Disclose

To assess the temporal resolution for cardiac motion using the TRI-PICCS (temporal resolution-improved prior image constrained compressed sensing [Med. Phys. 36(6), pp. 2130-2135]) and a TV minimization reconstruction algorithm in single source (SS) and dual source (DS) CT.

The TRI-PICCS method was proposed to reduce the data segment that is used to reconstruct a cardiac SSCT scan from 240° to 120°. The results shown were promising but limited to a single case. We generalized TRI-PICCS to DSCT and independently evaluated the temporal resolution of TRI-PICCS and TV minimization for SSCT and DSCT. The TV minimization reconstruction uses the TRI-PICCS algorithm with the fading parameter alpha being 0 such that no prior image is considered during reconstruction. We assess the temporal resolution as a function of the position in the field of measurement with respect to the source position. To do so, a high-contrast vessel, which performs a cardiac motion, is simulated at numerous positions. For reconstruction the rest phase of the motion is used. The vessel was simulated at ten different heart rates and the duration of the rest phase that corresponds to the maximum heart rate that still allowed an artifact-free reconstruction is considered as the temporal resolution at the given position. All simulations were performed four times using different motion directions in order to reduce the dependency on the motion direction. For the simulations the configuration of a state-of-the-art clinical DSCT scanner with 280 ms rotation time was chosen.

The results show that the temporal resolution is about 130 ms for the SS short scan, 98 ms for SS TRI-PICCS, 92 ms for SS TV minimization, 63 ms for DS short scan, 56 ms for DS TRI-PICCS, and 47 ms for DS TV minimization reconstruction.

Compared to a short scan TV minimization achieves highest temporal resolution results for SSCT and DSCT. Independently of the reconstruction method, the temporal resolution in SSCT is lower than a standard reconstruction of DSCT. However, our results are limited to artificial phantom data, with only one vessel moving, and it is not evident that clinical patient data will profit from these reconstruction approaches.

Improving temporal resolution in cardiac CT improved cardiac CT image quality and therefore allows for a more precise diagnosis of the diseases of the heart.

Maaß, C, Kachelrieß, M, Evaluation of Temporal Resolution-improved Cardiac Single Source and Dual Source CT Reconstruction Algorithms.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9014946.html