| Combinations of imaging modalities that integrate the strengths of two modalities, and at the same time eliminate one or more weaknesses of an individual modality, offer the prospect of improved diagnostics, therapeutic monitoring, and preclinical research. This explains the recent success of PET/CT and SPECT/CT systems that are playing an increasingly important role in the diagnosis and staging of human diseases. Research groups have been developing MR-PET scanners for small animal imaging. In addition, the first prototypes of a commercial MR-PET scanner for simultaneous human brain studies have been built. New work using MRI has shown that tracking of targeted molecules, agglomerates and cells, can be performed with sufficient sensitivity, albeit with compromised spatial resolution. Thus, MR-PET delivers an unforeseen spectrum of useful combinations for basic research and clinical applications.
The present state, expectations and challenges of MR-PET were discussed by 165 participants from 20 countries during an international workshop at the Forschungszentrum Jülich, Germany, held in conjunction with and following the 2008 IEEE NSS/MIC/RTSD at Dresden, Germany, on October 27-28, 2008.
During the introductory lecture Torsten Kuwert (Erlangen) stressed that MR-PET opens a totally new horizon of multi-parametric imaging by exploiting the functional capabilities of both MRT and PET. However, before this vision will be fully realized a number of problems have to be solved that relate to the hardware and the methods involved, as discussed during the two days. Volker Schulz (Philips) described his company’s two projects: a first non-hybrid MR-PET with the two modalities in line, and a later fully integrated hybrid scanner. Robert Krieg (Siemens) especially reviewed the functional features offered by future MRT in addition to its capability to image anatomy.
Bernd Pichler (Tübingen) gave a comprehensive overview of MR-PET scanner designs for imaging small animals. These ranged from solutions with long fibres connecting the PET detector and its electronics, which was detailed later by Jane Mackewn (London), through MR-PET with split MRT, as later presented by Rob Hawkes (Cambridge), or MR-PET with cycled field MRT, to MR-compatible PET-detectors based on solid state readout electronics instead of photomultipliers. The 7 T MR-PET constructed at Tübingen uses non-magneto-sensitive avalanche photo diodes (APDs) as readout electronics and shows negligible interferences between PET and MRT. Matthias Schmand (Siemens) described how a similar design has been realized in the first commercial MR-PET scanner, which inserts a newly constructed BrainPET in a 3 T MRT. Chris Thompson (Montreal) indicated how this system’s radial resolution may be improved by decreasing the depth-of-interaction (DOI) problem with dual layer crystals. This problem may be also circumvented by an LYSO-double-APD-based design suggested by Joan Varela (Lisbon), who plans to adapt a PET system dedicated for mammography (PEM) to work as a brain MR-PET.
Although APDs are found in current hybrid MR-PET, the lectures of Yong Choi (Seoul), Volker Schulz, Peter Bruyndonckx (Brussels) and Craig S. Levin (Stanford) made clear that silicon photomultipliers (SiPMs) offer better features so that many groups focus their research on this issue.
Dave Townsend (Knoxville) discussed the principal prerequisites needed for quantitative PET and that the different well-known corrections have to be adapted to the new detector technology of PET working in the MR environment. One primary correction is that for attenuation. As MR-PET does not offer a possibility for measured attenuation correction by transmission scans, new solutions must be found, as discussed by Elena Rota Kops (Jülich) and Gaspar Delso (München), who included this issue in his lecture on the advantages and challenges of whole-body MR-PET. The contribution by Johannes Breuer (Cologne) discussed how blood sampling necessary for quantitative PET modelling, as was later reviewed by Hidehiro Iida (Osaka), can be achieved within the MR environment.
The functional features of MRT are improved with an increasing magnetic field. The possibilities and challenges with ultra-high field MR-PET were finally presented by David Schlyer (Brookhaven), who plans to insert the RatCAP in a small animal 9.4 T MRT, and Jon Shah (Jülich), who directs the project of 9.4 T MR-PET for human brain studies.
Following the success of this workshop there was general agreement to have an update within the next two years.
For further information on this workshop, please email Organizing Committee members Hans Herzog and Karl Ziemons at mr-pet@fz-juelich.de or visit www.mr-pet-juelich.de.
|
