June 2 , 2008
PET/MRI: New Fusion
By Dan Harvey
Vol. 9 No. 11 P. 20
Software image fusion has been around for a while, but the desire for real-time fusion of MRI and PET images has hatched a new type of scanner.
Researchers have introduced a new modality coupling into the fusion technology arena. Working in Simon R. Cherry’s, PhD, laboratory at the University of California (UC), Davis, scientists developed an MRI-compatible PET scanner designed for studying mice.
Like other hybrid imaging techniques, this commingling fuses complementary capabilities. MRI provides superior soft-tissue contrast and high spatial resolution anatomical information, while PET offers high sensitivity. That description only simplifies the advancement and its potential, says Cherry, a professor and the chair of biomedical engineering at UC Davis. “From a research perspective, it goes well beyond that,” he says.
In addition to anatomical data, MRI brings into the marriage information related to perfusion and permeability, while PET can provide specific molecular information related to cell surface resectors, enzymes, and gene expression. As such, the PET/MRI combination represents a powerful research tool, Cherry adds.
The scanner uses Avalanche photodiode detectors instead of the photomultiplier tubes (PMTs) found in other PET scanners. Its design enables simultaneous rather than sequential MRI and PET scans, which means that patients do not need to be repositioned during imaging.
Explaining the importance of simultaneous acquisition in their paper, “Simultaneous In Vivo Positron Emission Tomography and Magnetic Resonance Imaging,” published in the March 11 issue of the Proceedings of the National Academy of Sciences, the researchers wrote: “To ensure that a subject is being imaged in the same physiologic state, and to correlate changes over time in the PET and MRI signals in response to an intervention … often requires that the data be acquired simultaneously. Simultaneous acquisition of PET and MRI data by using an integrated imaging device is, therefore, necessary to answer many important biomedical questions in dynamic, living systems.”
In the clinical setting, the fusion can potentially improve healthcare by increasing understanding of the causes, effects, and development of disease processes such as cancer and Alzheimer’s. It could also advance diagnosis and help monitor treatment, particularly related to cancer and bone disease. In addition, PET/MRI could help determine the efficacy of certain drugs by allowing clinicians to observe how the drugs travel throughout the body.
Additional benefits are that PET/MRI offers the potential to provide more structural detail than CT scans, especially when imaging soft tissue. And MRI does not expose the patient to ionizing radiation.
Advanced Research Projects
The new fusion underscores the broad and sophisticated intellectual and technological conceptualization that characterizes the focus of Cherry’s lab on the emerging molecular imaging field. Researchers actively engage in developing new and improved imaging technologies, contrast agents, and imaging probes and then apply these pioneering elements to molecular diagnostics and therapeutic agents.
Technologies and techniques the researchers work with include PET, MRI, SPECT, CT, and optical imaging. Lab members are particularly interested in creating new in vivo molecular imaging techniques and technology for the noninvasive rodent research model of human disease. Also, researchers have been developing PET technology for applications in breast cancer patients.
“We’re particularly excited about the dedicated breast PET/CT scanner,” says Cherry. “John Boone, PhD, a professor of radiology here at UC Davis, developed the scanner that produces three-dimensional imaging of the breast with x-rays at the same dosages as mammography.” The technology is currently in clinical trials, with several companies interested in the commercialization possibilities.
“The first patient study was done several months ago, and the images were exquisite,” says Cherry. “What’s most exciting is that we achieved much higher spatial resolution, both on the PET and CT of the breast, than could be achieved with a conventional scanner.”
Cherry adds that the technology could be especially useful for obtaining an earlier diagnosis and then monitoring treatment, particularly for high-risk patients.
The lab is also involved in improving molecular imaging, particularly high-resolution imaging of laboratory rodents, with microPET technology. “We’re trying to hit the PET resolution barrier because many people feel that PET isn’t a very high-resolution modality,” explains Cherry. “We are looking at just how good the resolution can be, and we think the answer is about a half-millimeter. So, we’re trying to develop an imaging system that will give us a half-millimeter spatial resolution in mice.”
Going from the ultra small to the other extreme, lab researchers are determining the viability of a longer whole-body PET scanner that could image the entire body in a single sweep. “Currently, you need to move a patient through a scanner in sequential fashion,” says Cherry. “We asked ourselves if it was possible to develop a scanner long enough to image the entire body at once and in a cost-reasonable way—and, if so, what would the images be like?”
What’s most intriguing is the potential capability to conduct whole-body dynamic studies. “A tracer is used and can be measured as a function of time, which opens the door for some fascinating pharmacologic-kinetic studies in collaboration with the pharmaceutical industry. However, right now all of that is at a very conceptual stage,” says Cherry.
From Concept to Reality
Cherry and his colleagues first conceptualized the PET/MRI fusion in 1996 and developed a rudimentary prototype one year later. In moving forward, they encountered several substantial challenges, not the least involving how PET and MRI systems interfere with each other and create artifacts and image distortion. The major concerns involved electromagnetic interference (EMI) and the negative impact that MRI’s magnetic field has on PET scanner detectors, which are typically based on scintillators coupled with PMTs. “These vacuum tubes are incredibly sensitive to magnetic fields and simply won’t work inside a magnet,” says Cherry.
Conversely, PET’s electrical and radiofrequency components can significantly disrupt the MRI system. To address this problem, the researchers deployed the silicon-based Avalanche photodiode detectors. “Obviously, we needed an alternative detector technology that would be much less impacted by magnetic fields. The magnetic field-insensitive photodiode detectors proved very effective. We used very short optical fiber bundles to appropriately place the photodiode detectors and PET electronics, with respect to MRI radiofrequency and gradient coils, to minimize interference. We’ve placed them in fields up to 9.4 Tesla without a problem,” says Cherry.
In this way, the researchers were able to develop a PET/MRI machine that effectively positions the PET scanner within the MRI scanner.
It’s interesting to note that with the early prototype, project participants used the PMTs. “However, we used them with very long optical fiber connections between scintillator elements and PMTs to bring the signal out of the magnet, which enabled us to position the PMTs somewhere else in the room,” Cherry says.
While this approach eliminated EMI between the PET and MRI systems, it proved impractical. PET scanner performance was poor compared with stand-alone PET scanners, and the length of the optical fibers made the system cumbersome. “So, from about 1997 to 2003, we waited for new detector technology to come along and mature to the point where it could make PET/MRI practical,” recalls Cherry. That happened when the Avalanche technology emerged in 2003. “That took us to the point where we could build the PET/MRI scanner. So the prototype we wrote about is really based on our last four years of work. We produced our first images in 2006 and have been using PET/MRI for biological research applications for the past two years,” he says.
UC Davis researchers are already seeking ways to improve on the PET/MRI fusion by increasing the spatial resolution and sensitivity of the PET component. Over the next several years, lab researchers will work on building a next-generation prototype that will provide better animal coverage and enhanced spatial resolution and sensitivity. “The current prototype works reasonably well and provides good images, and we’re collaborating with scientists on biological studies. But the new prototype will push performance well beyond current standards,” says Cherry.
He adds that lab researchers also will be developing imaging probes specific to the PET/MRI combination. “One of my colleagues has received funding to develop such probes. These are gadolinium probes, which are MR visible, and radioisotope probes, which are PET visible,” Cherry reports.
Clinical and Research Potentials
The evolving fusion technique has the potential to help rapidly develop new treatments. Cherry notes that PET/MRI already effectively evaluates new medical therapies in lab animal models, providing details about drug distribution and action that may help carry new drug therapies into the clinical setting.
Looking further down the road at clinical applications, researchers believe that PET/MRI may lead to a better understanding of neurological conditions such as Alzheimer’s disease, Parkinson’s disease, schizophrenia, and epilepsy. It could also prove beneficial in the treatment of stroke patients because it may help identify recoverable areas of brain tissue.
As far as research potentials, Cherry sees many opportunities to use PET/MRI. “Many complementary opportunities exist. All you need do is look at the literature about the many applications—such as cardiovascular, cancer, and neuroscience—where people have applied PET and MRI separately. Now, they’d be able to do simultaneous scans. That will be very valuable in studies where aspects change with time. PET/MRI enables you to capture a snapshot. That’s one of the ways the combined device is most valuable.”
Cherry also foresees PET/MRI’s value in stem cell therapy research. As the fusion technique enables simultaneous measures of anatomy, function, and biochemistry, it could facilitate the tracking of stem cell migration into damaged brain regions, for example. “Cell tracking represents just one of the most interesting research areas, as PET/MRI allows researchers to monitor the stem cells. I think this research direction will be particularly fruitful.”
Innovative introductions are usually followed by questions about when the development will be commercialized. It’s hard to say when PET/MRI will move to the mainstream clinical setting.
“Siemens has developed a prototype for distribution in the United States as a research tool,” reports Cherry. “But they’re not ready yet to offer it as a product.” Siemens Medical Solutions, Germany introduced the PET/MRI prototype in May 2007 in Berlin at the International Society for Magnetic Resonance in Medicine annual meeting. The system has been used for in vivo human brain imaging studies at Siemens’ U.S. medical facilities. Complete testing on the system began in late 2007.
“I’ve also heard that other vendors are busy working on PET/MRI,” adds Cherry. “When major medical companies invest money into technology, you know that products are imminent.”
— Dan Harvey is a freelance writer based in Wilmington, Del., and a frequent contributor to Radiology Today.