There are many motivations for adding simultaneously acquired MR images to PET scanning. The most straight forward are, superior registration of MR and PET images, the addition of morphological detail when there is non-rigid motion and for pre-clinical studies simultaneous imaging could lead to a significant reduction in the time that animals are required to be anesthetised. In addition simultaneous MR has the potential to provide accurate motion correction for PET image reconstruction. For functional imaging simultaneous acquisition is required to assess the subject in the same physiological state, such as acute stroke studies. The elimination of the additional radiation associated with combining CT with PET, by providing anatomic detail with MR, would be a crucial advantage for cancer screening. Combining the two instruments necessitates some engineering tradeoffs, especially associated with the use of the highly developed photomultiplier tube (PMT) used for light amplification, because of its incompatibility with strong magnetic fields. Our approach is to provide a split in the magnet and gradients to locate the magnetic sensitive components, the PMTs, in regions of low magnetic field, leaving only the essential PET components, the scintillator blocks, in the strong magnetic field region. The crystals are coupled to the PMTs by extending the optical fibres. A further advantage accrues by moving the PET electronics out of the region seen by the MR radio-frequency (RF) and gradient coils as electromagnetic interference effects between the PET and MR systems, which could cause artefacts in either modality, are eliminated. Here we describe a preliminary evaluation of the system, which is essentially a microPET Focus-120 located in a 1T split magnet, and compare its performance to previous microPET instruments.

Key words: PET; MRI; positron emission tomography; magnetic resonance imaging; multimodality imaging.