7T scanners move closer to use in clinical research

September 1, 2005

Philips and Siemens are delivering on promises made two years ago to develop ultrahigh-field MR systems. The companies are completing construction of several 7T clinical installations, each bearing the clinical front end of a mainstream MR scanner rather than the unwieldy controls that have marked previous installations. The ramifications for the MR community are enormous.

Philips and Siemens are delivering on promises made two years ago to develop ultrahigh-field MR systems. The companies are completing construction of several 7T clinical installations, each bearing the clinical front end of a mainstream MR scanner rather than the unwieldy controls that have marked previous installations. The ramifications for the MR community are enormous.

These types of operator interfaces promise quick and relatively painless translation of results from 7T to the growing installed base of 3T scanners. Their implementation and the pathfinding data they are expected to yield could help vendors fulfill their promise of advanced applications at 3T, which is a major reason customers buy 3T scanners.

The clinical goal for research conducted at 7T will be to gain a better idea of what may be achieved at lower fields. Structures and conditions revealed at ultrahigh field strength might be found at 3T if physicians knew where to look. Engineers could realize this possibility if they can rework the protocols.

A showcase for 7T opened in June at the Philips MRI research facility in Highland Heights, OH. The corporate facility is a demonstration and training ground for physicians and scientists interested in obtaining ultrahigh-field scanners. The $10 million Achieva 7T system operating there is similar to one being installed at the Wright Center of Innovation in Biomedical Imaging at Ohio State University.

Research conducted on the Philips-operated Achieva 7T indicates that the increased susceptibility effects present at this field strength can be exploited to produce high-resolution vascular images, including black blood venous imaging. Three-D magnetization prepared turbo field-echo imaging demonstrates small anatomic structures in the brain, such as peduncle striations, and exceptional gray-white matter contrast at high resolution, revealing subtle anatomy such as white matter tracts. Studies at the Philips center have demonstrated the scanner's ability to conduct single-voxel spectroscopic imaging of metabolites, including peaks of N-acetylaspartate, creatine, and choline.

John L. Patrick, Ph.D., Philips' director of business development for MR, expects that the company will install 15 to 20 7T whole-body clinical units. The purpose of their development is to expand the frontiers of MR imaging while establishing Philips as a provider of leading-edge technology.

"We don't sell these systems to make a profit on them," Patrick said. "We are spending in excess of $20 million in development on 7T, and even at the wildest price margins, we would never recover our costs."

Philips has contracted to install two units in the U.S., including the OSU system, and two in Europe. Siemens has installed one 7T scanner in Europe and two in the U.S., and it has orders for eight additional clinical units. A 7T system began operating in late spring at New York University, where researchers plan to look particularly at the brain, but also the breast, ovaries, and prostate. Much of the work so far has been preliminary, as the research team is just beginning to work out the different imaging protocols.

"You appreciate how far modern clinical scanners have come, because you realize this is how people started doing MR way back when nothing was written," said Dr. Vivian Lee, a professor and vice chair of research at NYU School of Medicine.

Most of these ultrahigh-field systems are in the hands of researchers who don't mind doing a lot of tweaking and tuning to get what they want. Dr. Tsutomu Nakada, for instance, was recruited from the University of California, Davis (where he still holds a professorship in neurology) to build a research institute for ultrahigh-field MRI in Japan. The Center for Integrated Neuroscience at the University of Niigata has been running a human 7T system and a 3T system at for about a year.

Nakada, director of the center, believes the main advantage of 7T is its higher signal-to-noise ratio, which is beneficial primarily in MR microscopy and MR spectroscopy.

Preliminary studies on the 7T scanner, built by GE Healthcare, have achieved 80-micron resolution in the living human brain. Nakada expects to obtain 40-micron resolution within the year but is hoping for 20.

"What we can do with this resolution is in vivo pathology, such as detecting senile plaque in Alzheimer's patients," he said.

Nakada expects that 7T MR will provide a major boost for MR spectroscopy-a welcome respite from the current state of affairs.

"To date, MRS in humans is sort of a joke," he said. "We could do it seriously with 7T or higher systems."

GE Healthcare has built several 7T systems. Its latest, installed but not yet operating at the University of Pittsburgh Medical Center, is outfitted with a clinical interface compatible with those on the 3T and 1.5T units running at the center. Fernando Boada, Ph.D., director of the MR Research Center at the University of Pittsburgh, looks forward to translating experimental work done at ultrahigh field to a clinical scanner.

"Our 7T scanner will be a good test bed on which to optimize 3T applications," he said.

Boada, a professor of radiology and bioengineering, expects the ultrahigh-field system to reduce scan times by a factor of four. The time saving will afford the opportunity to develop techniques that can be built into more efficient protocols at 3T.

"This will have great implications for cancer," he said. "There will be many ways of looking at cancer that we can translate for use at 3T, where we can try to detect response to therapy or identify the metabolic signatures of different kinds of cancer. Right now, I can do these applications on 3T, but they are very hard to optimize."

The Philips 7T scanner at OSU, scheduled to begin operation by year-end, will support research in cancer, neuroscience, and cardiovascular disease, providing high-resolution images of angiogenesis and improved imaging of blood flow and oxygen use in the brain. It is intended to be used particularly for study of degenerative neurological diseases, including Alzheimer's, Parkinson's, and multiple sclerosis.

Dr. Michael V. Knopp, chair of radiology at OSU, views MR field strengths as a continuum along which capabilities pioneered with animal scanners must be translated for use on human systems. Knopp and his colleagues have been operating an 8T human system and a 9.4T animal system for several years. The university is installing the clinically oriented Philips 7T system adjacent to an operating clinical 3T unit.

"The idea is to have a cross-correlation among the field strengths," Knopp said.

Advances in parallel imaging have resulted in higher resolution, which the Ohio team hopes to use for identifying patterns at 7T that can be found, with increased resolution, at 3T and even 1.5T.

"I think ultrahigh-field MR will be the facilitator to drive innovation, which, with further technical advances, can be pushed into lower fields," he said.

Peter Morris, Ph.D., director of the Peter Mansfield MR Center at the University of Nottingham in the U.K., expects to have a Philips-built 7T scanner running by late summer or fall. Preparatory clinical work is under way on a 3T unit that Morris and his colleagues built and have operated for the past 13 years.

"We thought building our own 7T was a bit too far a reach," he said.

The need to explore MR functional imaging and spectroscopy impelled the Mansfield Center to embrace 7T. Advanced capabilities possible with ultrahigh-field imaging may allow diagnosis of schizophrenia. Morris hopes to capture a schizophrenic episode with 7T MR.

"In the past, we have averaged those activities over a number of episodes, so we lose the possibility to distinguish the uniqueness in each of them," he said.

The same need to average data blurs the results of functional MRI. Actions such as grasping an object must sometimes be repeated as many as 30 times to visualize brain activity. This averaging obscures the subtle changes that may accompany each action.

Future technological developments constitute the great unknown for MR. New contrast materials that enhance specific structures or functions might provide an immense boost to clinical 3T and 1.5T imaging. Developing such agents, however, hinges on identifying clinically relevant targets that might be visible only at ultrahigh fields. The impetus to achieve this breakthrough will drive developments by Knopp and his OSU colleagues in their studies at 7T.

"With the combination of different developments in nanotechnologies, using the higher sensitivity they confer and creative sequence design, there will be a lot of clinical potential to explore at lower fields," he said.