Edwin ter Voert, Presenter: Nothing to Disclose

Gaspar Delso PhD, Abstract Co-Author: Employee, General Electric Company

Sangtae Ahn, Abstract Co-Author: Employee, General Electric Company

Florian Wiesinger, Abstract Co-Author: Employee, General Electric Company

Martin W. Huellner MD, Abstract Co-Author: Nothing to Disclose

Patrick Veit-Haibach MD, Abstract Co-Author: Research Grant, Bayer AG Resaarch Grant, F. Hoffmann-La Roche Ltd Research Grant, General Electric Company

In PET/MR imaging artefacts due to (metal) implants could lead to inconsistencies in the MR based attenuation maps and affect the PET images. In this work we evaluated the influence of TOF on image quality in TOF PET/MR scanning by comparing TOF and non-TOF PET reconstructed images.

In this prospective study, 25 patients were so far evaluated in a new, simultaneous whole-body TOF PET/MR scanner. To assess the differences between TOF and non-TOF, all PET images were reconstructed twice: with and without the TOF information. A fully 3D-OSEM iterative reconstruction algorithm, incorporating all corrections (scatter, random, dead time, attenuation, and normalization), was applied. All datasets were based on the standard Dixon-based attenuation correction sequence. The images were analyzed by a dual-board-certified radiologist/nuclear medicine physician with 11 years of experience. Image quality was scored: 0) no artefact, 1) non-significant artefact, 2) clinical significant artefact but reader confidence not impaired, 3) clinical significant artefact with reduced reader confidence, 4) pathology missing.

So far, in 9 patients metal artifacts were discovered on the MR-AC. In 1 patient with spondylodesis the TOF images scored 1, and non-TOF images 4 as a bone metastasis was not seen. In 3 patients with a port, TOF scored 0 and non-TOF 1. One patient with a right basal lung artefact scored 0 on the TOF and 2 on the non-TOF images based on the size of the artifacts. Two patients with artifacts in the upper abdomen scored 1 and 1 in the TOF images and 1 and 2 in the non-TOF images. Two patients with artifacts in the lower abdomen scored 1 and 2 in the TOF images and 1 and 3 in the non-TOF images. Cardio artefacts in two patients had the same scores (1/2) in both the TOF and non-TOF images, although the TOF images were found to have improved overall image quality. In patients with dental implants/fillings, stents and intrauterine devices, no (significant) artefact was seen.

The PET/MR reconstructions benefit from the TOF information as it reduces the artefacts especially around metallic implants and might even recover pathologies which can be missed on the non-TOF PET-images.

Inconsistencies in MR based attenuation maps could result in artefacts in PET imaging. Lesions near implants may therefore be obscured. TOF-PET/MR significantly reduces these artefacts near implants.

ter Voert, E, Delso, G, Ahn, S, Wiesinger, F, Huellner, M, Veit-Haibach, P, The Effect of TOF on PET Reconstructions in Patients with (Metal) Implants in Simultaneous TOF PET/MR Scanning.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14010777.html