CONTACT: Stanford University News Service (650) 723-2558

Eye test devised to measure moving picture's path through brain

STANFORD -- A group of Stanford researchers has pushed imaging of the human brain to a resolution almost 50 times greater than previously accomplished.

Their magnetic resonance imaging technique for distinguishing neural activity at locations separated by as little as 1.4 millimeters in the brain is outlined in the June 16 issue of the journal Nature.

The more detailed images of the primary visual cortex of the brain were achieved not by improvements in magnetic resonance hardware but by cycling a brain stimulus in a way that allowed an ordinary clinical machine to track blood flow at a more detailed level.

"We applied our ideas to vision, but we think we see how to use these same methods for other modalities," such as hearing, tactile sensation and perhaps eventually even language and memory, said Brian Wandell, professor of psychology.

Wandell, whose specialty is visual perception, conducted the research with graduate student Stephen Engel and Professor David Rumelhart of Stanford's Psychology Department, and with Adrian Lee and Gary Glover of the Department of Radiology and Michael Shadlen of the Department of Neurobiology at the Stanford School of Medicine.

Their brain mapping work demonstrates the potential for imaging to aid diagnosis and treatment of a wide range of sensory problems that are not well understood. These might include temporary or permanent damage from strokes or accidents, as well as developmental problems, where it would be helpful to distinguish and locate very specific brain functions within normal humans and in individuals with some loss in brain function.

MRI technology is improving and decreasing in price, Wandell said, so that every medical clinic is likely to have an MRI machine in the future. Micro MRI machines that are being demonstrated now allow surgeons to work inside them, taking images of the area of the body upon which they operate as they are cutting with special plastic tools.

An unrealized hope for MRI, however, is that it can some day pinpoint which neurons are involved in a specific brain-controlled function. By tracking blood flow, magnetic resonance has allowed researchers to see brain activity on the level of millions of neurons, Wandell said, but previous attempts to see it on the level of a few hundred thousand or fewer have been frustrated by surges of blood flowing through blood vessels to reach specific neurons.

Wandell's research team attempted to improve MRI resolution in a segment of the primary visual cortex, where the functions are already well known from crude mapping techniques, such as war-time studies of the vision damage to soldiers with head wounds. A key to improving resolution was devising an eye chart, or moving image, for the person to watch while having his or her brain imaged.

"We wanted an image that would not cause sudden surges of blood to the entire primary visual cortex region we were imaging," Wandell said. The moving stimulus they settled on is a videotape of concentric circles of white alternating with a checkerboard pattern. Like a child's pinwheel being spun with precision, the checkerboard circles regularly move in and out across the field of vision, so that the neurons responsible for monitoring different locations in the field respond only when the flickering checkerboard passes into the degree-range that they monitor.

"In this stimulus, at any given time, half the visual field is stimulated and half is not," Wandell said. "As the image moves, it just varies on which half is on or off, so the total blood flow to the area is roughly constant, and we don't have these big surges of blood in and out. That enhances our resolution and allows us to see the image cycling through the cortex."

While the researchers can't see down to the level of individual neurons, they are able to measure differences in blood flow in cortical locations separated by less than 1.4 millimeters along continuous strips of the visual cortex. Other brain imaging methods, including the more invasive positron emission tomography, have typically achieved spatial resolution in the range of 7 to 10 millimeters and are able to do somewhat better by averaging results over a number of patients tested. To be useful in clinical work, however, a method has to reliably measure each individual.

Since the brain is a three-dimensional object, the improvement achieved in measurement of a linear strip of it actually represents a 50- fold improvement in ability to image the brain's volume.

"There's nothing discretely different about what we did from other MRI researchers," Wandell said, "but we were able to show that you can use a periodic stimulus to generate periodic waves of neural activity."

A proper stimulus, he said, would need to be developed for each sensory modality that researchers wanted to study. In hearing, for example, such a cycling stimulus would take advantage of knowledge that the auditory areas of the brain are laid out from high pitch to low pitch. In cases like memory and language recognition, researchers probably would first need to have a better idea of the brain's organization to design stimuli based on this concept, he said.

The Stanford team is particularly interested in applying its technique to better understanding of the visual developmental processes that relate to the plasticity, or the flexible and inflexible patterns, of the brain, Wandell said. There is mounting evidence that the brain reallocates cortical resources so that, for instance, when someone loses the use of an eye or a limb, the neurons temporarily or permanently go to work for some other body function. This is believed to be what is happening with "phantom limbs" - reports from people who have had a limb amputated that they feel the sensation of having that limb touched when some specific other location on their body is touched instead. Better understanding of the brain's plasticity, Wandell said, could be used to devise treatments for a wide variety of conditions.

In vision, for example, he said, a "wandering" or "lazy" eye is often treated by patching the good eye so that the muscles of the bad eye are forced to develop strength.

"But that means the good eye is in disuse, which may cause the brain to reallocate cortical resources. Part of your visual cortex may take up a new function" and the good eye may never fully recover its past capabilities.

"What will be really exciting," he said, "is when, as a result of these images, we think we can figure out exercises, electrical stimulation or something that will help somebody who has lost their vision."



This is an archived release.

This release is not available in any other form. Images mentioned in this release are not available online.
Stanford News Service has an extensive library of images, some of which may be available to you online. Direct your request by EMail to newslibrary@stanford.edu.