A Meeting of Minds: Using Computers To Study The Brain

Sidebar:The Brain: Byte By Byte Given that computers were first invented with the intention of duplicating certain functions of the brain--notably memory and calculation--researchers say it is perhaps appropriate that today, science has evolved to the point at which the machines are being used to study that organ, employing the selfsame properties they emulate. "Except for [a computer's] great memory and speed, the brain is much more powerful," says Stephen Koslow, director of the Division of Neuroscience and Behavioral Science at the National Institute of Mental Health in Bethesda, Md. "A computer is a more rigid system, but it is sort of interesting how things have come to a full circle." For more information on neuroscience and brain research, contact: Society for Neuroscience 11 Dupont Circle, Suite 500 Washington, D.C. 20036 Phone: (202) 462-6688 Nancy Beang, executive director Larry Squire, president 23,000 members American Academy of Neurology 2221 University Ave., S.E. Suite 335 Minneapolis, Minn. 55414 Phone: (612) 623-8115 Fax: (612) 623-3504 Jan Kolehmainen, executive director Jack Whisnant, president 12,800 members Scientific interest in studying the brain and nervous system has boomed in the past few decades--as evidenced, for example, by the dramatic increase in the membership of the Washington, D.C.-based Society for Neuroscience, from 500 in 1969 to 23,000 in 1994. The public's interest in neuroscience has also grown because of the rising awareness of disabling diseases like Alzheimer's and Parkinson's, stroke, and depression, which is especially reflected in the large number of private organizations--such as the Alzheimer's Association in Chicago; the Arlington, Va.-based National Alliance for the Mentally Ill; and the Tourette Syndrome Association in Bayside, N.Y.--that support research in specific areas of neuroscience. Expounding on the applications of research to treating brain disorders, in 1988 the late Rep. Silvio Conte (R-Mass.) introduced a bill in the House of Representatives to promote neuroscience. This resulted in an official proclamation, signed by President George Bush in July 1989, declaring the 1990s as the "Decade of the Brain." It calls upon all public officials and people to observe the decade with programs and activities appropriate to the theme. One of the initiatives to arise from the Decade of the Brain has been the Human Brain Project, a broad-based federal effort, supported by several agencies (see accompanying story), to use computers in mapping the human brain. "What we are attempting to do is take advantage of the advances in computer and communications technologies to develop tools to study [the brain and nervous system] at different levels and to allow various researchers to share information with each other," says Koslow. "Neuroscience is such a vast field," says Frank Eeckman, a neuroscientist at Lawrence Livermore National Laboratory (LLNL) in Livermore, Calif. "There are several groups working in different aspects of neuroscience research-- genes, neural circuits, gross anatomy, and psychology--and there is such an enormous scale of difference in their work that they can't talk to each other." While the Human Brain Project was formalized only last year, various neuroscientists are already using computers to facilitate their research. "A vast array of computer-based tools is needed to complement the diverse varieties of data obtained about the brain," says Peter Fox, director of the Research Imaging Center at the University of Texas Health Science Center in San Antonio. "There are categories of software depending on what you want to do," he says. "There are devices and sensors which monitor brain activity by various criteria--blood flow, volume, or oxygen levels--and we need different types of software to get the different types of data. These [programs] are acquisition software which tell the machines how to acquire the data. Once we have the data we need additional, post-processing software to use the information for statistical and other analyses." Using these various tools, Fox and his colleagues developed one of the first functional online databases--called Brain Map--to correlate structure and function of the brain. Set up in an interactive, multimedia format, BrainMap contains numerical and statistical data from peer-reviewed journal articles, as well as pictorial information digitized from the Atlas of Stereotaxic Anatomy of the Telencephalon (J. Talairach et al., Paris, Masson & Cie, 1967). "We've incorporated fairly elaborate indices for accessing the information," says Fox. "Search categories include anatomy, behavior [function], citations, and methodologies used." "If a scientist were interested in a particular aspect of vision research, he or she could search the database under that term," explains Jack Lancaster, a physicist at San Antonio who worked with Fox in setting up BrainMap. "The computer would then come back with the number of entries-- articles, experiments, and locations--in the database corresponding to the search criteria. The researcher would thus know the nature of the experiments, who did them, and what the findings were." The embedded atlas provides a pictorial guide whereby the researcher can call up images and look up where the research findings are located within these images, explains Lancaster. "When a user calls up an image, the computer screen provides three images--a large diagram representing a slice of the brain in one plane, and two smaller transparent views of the brain in the two perpendicular directions to help pinpoint the locations of various points in the main diagram," he adds. "The database would have an entry capability, as well, to enable scientists to enter new data and use previous publications to interpret their new observations," says Fox. Currently BrainMap's software is in beta test-phase, which means that a limited number of users are evaluating its capabilities and limitations. Fox and his team hope to have it widely accessible sometime next year. Because no two brains are alike, a key element for the success of the database, say BrainMap's creators, is ensuring that structural data from different sources be set to a standard scale and that each point in the brain be made identifiable by a coordinate system. "When comparing two images, we have to make sure the two brains are in register and that the corresponding parts are in the same place," LLNL's Eeckman says. "For example, an MRI [magnetic resonance imaging] and PET [positron emission tomography] scan from the same brain need to be in register to correlate the information from the two experiments, otherwise the comparisons are meaningless." David Van Essen, a neurobiologist at the Washington University School of Medicine in St. Louis, has developed tools such as warping algorithms to "fit" one brain to another so as to enable meaningful comparisons. These algorithms endow researchers with powerful tools: Physicians, for example, could compare the brain of an ill patient to that of a normal person, say researchers. Tools like the warping algorithms also have applications in brain modeling, says Van Essen, who has done extensive work on mapping the visual cortex of the macaque brain. Researchers can thus apply the insights gleaned from the brain in one system to model the working of the brain in another system. At the Scripps Research Institute in La Jolla, Calif., Floyd Bloom's team has already digitized atlases of the brains from other animals such as cats, rats, mice, and monkeys. While the existing databases are prototypes, Koslow anticipates that the final product of the Human Brain Project will be a much larger and a far more powerful system. "Calling it a map or even a database is a misnomer--the terms are too limiting," says Eeckman. "A data-management system would be a more appropriate description." "At this point we haven't determined the final configuration, and it will probably be half-a-dozen years before we have some inkling of what it will become," says Koslow, taking into account the ongoing developments in computer science (for example, three-dimensional virtual reality technology) that would influence the project. "Once [the database] is up, scientists should be able to access it from their terminals, either to retrieve or feed information, and use it for modeling work. We anticipate it will have multiple uses--not only in research but in education, therapy, and intervention, as well."

A Meeting of Minds: Using Computers To Study The Brain

Interested in reading more?

Become a Member of