Project Summary/Abstract
We propose the clinical translation and validation of an innovative “radiofrequency (RF) penetrable”
dedicated positron emission tomography (PET) insert technology that can be placed within any stand-
alone magnetic resonance imaging (MRI) system for acquiring simultaneous PET/MRI data. For this
proposal the Stanford team will work with industry partner “PETcoil” to address technical developments
required for translational studies, ensure industry level standardization of end user software and
technology, and acquire first in-patient simultaneous PET/MRI data using this novel PET insert. PET/MRI
exhibits some attractive features. First, combined PET/MRI is uniquely capable of providing excellent
anatomical soft tissue contrast and multi-parameter information in a single a scan and, as a result,
PET/MR is now commonly used for characterizing disease in regions such as the brain, head and neck,
breast, liver and pelvis. Second, MRI does not introduce ionization radiation in a PET/MR study. This
significant reduction of radiation dose makes PET/MRI an attractive modality for pediatric patients and
those requiring recurring PET studies. Third, PET/MR imaging can occur simultaneously, unlike
sequential PET/CT, resulting in temporal in addition to spatial correlation of PET with MR data. Despite
these attractive features, the adoption of PET/MRI has been slow since its introduction in 2011. One main
reason for this slow adoption is the high cost of procuring an integrated PET/MRI system, which is about
$5M for the machine and another $1.5–2M for required room renovations to host the machine. The
resulting cost (~$6-7M) is simply not affordable for most institutions. Furthermore, the permanently
integrated PET/MR system designs offered by vendors such as GE and Siemens yield sub-optimal spatial
resolution and sensitivity performance for the PET system component, especially for neurological imaging
applications. The proposed brain-dedicated PET ring offers higher photon sensitivity and spatial
resolution, and can be inserted into any existing MR system; thus a user only needs to procure the insert,
and MR system modifications or room renovations are not required, reducing the entry costs roughly 10-
fold. We have realized a 1st generation brain dedicated PET insert for simultaneous PET/MR and are
completing a 2nd generation version, both with a spatial resolution of 2.7x2.7x2.7 mm3 and the latter
achieving a photon sensitivity of >6%. In addition, this insert approach is “RF penetrable,” an important
and novel concept enabling collection of PET/MR data using the MR system’s built-in body coil for RF
transmission through the PET ring insert into the patient, and only a RF receiver coil resides inside the
insert, thus facilitating the lower cost insert concept for achieving PET/MR. We will translate and validate
the 2nd generation brain PET insert and work with our industry partner to standardize our data processing
workflow. If successful, this project will enable more widespread dissemination/accessibility of PET/MRI.
