Meeting Banner
Abstract #2810

Feasibility of PET Attenuation Characterization of MR Hardware Using 3D Data from a LINAC as Radiation Source

Stephan Biber1, Ralf Ladebeck1, David Faul2, Yvonne Candidus1

1Siemens Healthcare Magnetic Resonance Imaging, Erlangen, -, Germany; 2Siemens Healthcare Molecular Imaging, Knoxville, TN, United States

With the introduction of integrated MR-PET systems both for preclinical animal studies as well as full sized human scanners, the question of PET attenuation originating from MR hardware (MR-HW) which is located between the patient and the PET camera, has been a topic of increasing interest for research. Prvious studies have identified mechanical structures for the patient table, the housings of local coils as well as the electronics and antenna structures within these housings as the major contributions to PET signal attenuation. This attenuation map needs to be included in the PET reconstruction. The goal of this paper is to evaluate the feasibility and the potential advantages of MR-HW attenuation correction based on 3D data coming from a LINAC instead of a CT, which has been used for measurement of hardware attenuation in earlier studies. The advantage of a LINAC based scan can be the reduced level of artifacts, as the LINAC provides a radiation beam with photons of much higher energy than the CT.

Keywords

accelerating acceleration according accurate addition allows almost antennas apertures applied appropriate array artifacts artiste attenuation beam bigger bilinear bone bottom bremsstrahlung cable cables canner cause challenging characteristics characterization clearly closer coil coils comes coming complete compute consisting contributions conversion copper correction corrections count criteria curve definition despite detuning difficult dynamic electronics energy especially even exact experiment extended extends feasibility forward full furthermore generating goal hard hardware head housing housings human humans included increasing instance integral integrated intersects larger linear local longer makes making maps material metal metals molecular much neck needs neglecting neighborhood nevertheless nuclear outlook paper part partial patient performance photon positioned predict pronounce quantitative question radiation rather rays reconstruction relies require required rescaled response sander scanners scatter science simulations slope soft solder sources spectral straight streaking strong strongly structures studies subject systems table target thorax tissue together transformation trap tube typical units upper variable vicinity voltage voltages windowing yield