Conventional MRI (right) andDR-MRI (left) in a patient with a ganglioglioma displacing white matter fiber tracts. Only the DT map localizes individual fiber tracts, allowing their preservation during tumor resection Credit: Courtesy of Aaron Field

Last month saw additional evidence that the National Institutes of Health (NIH) is increasing its activities in the area of biomedical engineering and bioimaging. The National Institute of Child Health and Human Development (NICHD) has signed a major deal with GE Medical Systems, licensing them to produce and market a new imaging technology it has developed. The new method gives high-quality views of nerve-fiber pathways, blood vessels, skeletal muscle, heart and other soft tissues, allowing the construction of detailed three-dimensional tissue maps.

Called diffusion tensor magnetic resonance imaging (DT-MRI), the system produces images noninvasively and painlessly. For certain scans, conventional MRI uses exogenous contrast agents, such as an intravenously administered gadoliniun-based compound. In contrast, DT-MRI exploits the effects of tissue microstructure on the diffusion of water to generate endogenous tissue contrast scans.

The motion of water molecules in tissues varies between tissues and according to their disease state. For example, in brain gray matter, water diffuses in an approximately spherical pattern, whereas in tissues with a large number of parallel fibers, such as brain white matter, skeletal and cardiac muscle, water diffuses fastest along the direction of the fibers, and slowest at right angles to them. Based on these differences, DT-MRI produces intricate 3-D images of the tissue's architectural organization and local structure. Changes in tissue water diffusion can be correlated with processes that occur in development, degeneration, disease and aging.

According to DT-MRI's principal inventor, Peter Basser, “The most important clinical application of DT-MRI to date has been the ability to follow the progression of cerebral ischemia during acute stroke, and to follow the subsequent neural degeneration in chronic stroke.” Basser, NICHD's Chief of the Section on Tissue Biophysics & Biomimetics, says “the first applications are likely to be in surgical planning, since DT-MRI can provide the neurosurgeon with additional information about brain anatomy and architecture that conventional MRI methods do not.”

Indeed, DT-MRI's use in imaging fiber tracts may cause a modest revolution in neurosurgery. Aaron Field, assistant professor of Radiology at the University of Wisconsin Medical School, is exploiting this property of DT-MRI to study cerebral tumor infiltration. His group has found that DT-MRI patterns indicate whether a tumor is displacing, invading or destroying white matter tracts. “Diffusion tensor properties differ between tumor types and even between different regions of the same tumor. We've used DT-MRI to map the displacement of functionally critical white-matter tracts, such as the corticospinal tract, by an enlarging tumor, enabling the surgeon to preserve the tract during resection.”

Massimo Filippi, director of the Neuroimaging Research Unit at the Scientific Institute and University in Italy, is using DT-MRI to study the brains of patients with multiple sclerosis (MS) and Alzheimer disease (AD). In AD patients, for example, Filippi's group found cortical gray matter abnormalities that correlate with the patients' clinical symptoms: “Diffusion changes in white matter (corpus callosum, white matter of the frontal, temporal and parietal lobes) are present in AD patients and are strongly correlated with mental state examination score.”

The only drawback to the technique appears to be artifact production. “DT-MRI data suffers from artifacts as do other imaging modalities,” says Basser. “The primary sources of these are background noise, image distortion due to eddy-currents, and rigid and non-rigid patient motion.” But Basser insists that DT-MRI is still an improvement over conventional MRI in this regard.

The NIH's interest in biomedical imaging and engineering has grown visibly since 2000 when Congress pronounced this an area in which the Institutes' efforts have been “inadequate” (Nature Med. 6, 7; 2000). Congress called for the creation of an Office of bioimaging/bioengineering that has now become the newest, full-blown Institute, called the National Institute for Biomedical Imaging and Bioengineering (NIBIB). This month, Roderick Petigrew takes over as NIBIB's first permanent director. In addition, Congress chose to elect Elias Zerhouni, an expert in MRI, as the new NIH Director.