Accuracy of MR-based attenuation correction First experience with a whole-body PET/MR system

Purpose/Introduction: Sequential PET/MR scanners for whole body investigations have been developed only recently and are starting to become commercially available. In 2011, one of the first of these combined systems (Ingenuity TF PET/MR, Philips) was installed at our institute. PET/MR is expected to offer new possibilities, in particular in the field of quantitative bimodal functional imaging [1]. Quantitative PET requires attenuation correction (AC) which is commonly based on a measurement of the photon attenuation using either a radioactive source in standalone PET (RSAC), or CT in PET/CT systems (CTAC). In PET/MR systems such a measurement cannot be performed, therefore AC is based on a dedicated MR measurement (MRAC) plus segmentation and tissue type identification [2]. Here, we report on a first evaluation of MRAC in whole body investigations with the new system.

Subjects and Methods: Evaluation of MRAC is performed by a direct comparison of MRAC maps acquired with the Philips Ingenuity TF PET/MR and RSAC maps acquired with a Siemens ECAT Exact Hr+ PET scanner for 18 patients. Moreover, the quantitative influence of AC on the reconstructed images was investigated.

Results: In most cases the MRAC maps match the RSAC maps very well. Artefacts due to dental implants and sternal cerclages were found. In one case, the algorithm was not able to separate tissue from air in the patients lung. The influence of these artefacts on the reconstructed PET image will be reported.

Discussion/Conclusions: The vendor-provided MRAC algorithms yielded good results in our first patient measurements with respect to soft tissue and air segmentation. However, the algorithm relies on anatomic reference data and thus artefacts arise if the anatomy of the patient does not fit to the reference. In the future, an improved separation between tissue, bone and air without using anatomic reference data might be achieved with ultra short echo time sequences [3,4].

References:

[1] Pichler, B.J. et al., 2008, Sem. Nucl. Med., 199-208
[2] Martinez-Möller, A. et al., 2009, J. Nucl. Med., 520-526
[3] Catana, C. et al., 2010, J. Nucl. Med., 1431-1438
[4] Keereman, V. et al., 2010, J. Nucl. Med., 812-818