Back To News>>

11/17/2006
SCCN at UCSD Receives NSF Grant for Study of Multimodal Dynamic Imaging of Human Brain Activity

Principal Investigator Scott Makeig and co-investigators Profs. Bhaskar Rao, Kenneth Kreutz-Delgado of Electrical and Computer Engineering, University of California San Diego (UCSD), together with Swartz Center for Computational Neuroscience (SCCN) postdoc Rey Ramirez and students of the co-investigators recently received a National Science Foundation (NSF) grant for collaboration on computational modeling of human EEG and MEG data. Here is a summary of the new results that formed the basis of their proposal:

"The experimental goal of cognitive neuroscience is to image human brain activity non-invasively with the highest possible space and time resolution. Electroencephalography (EEG) and magnetic EEG or magnetoencephalography (MEG) have a remarkable temporal resolution, recording complex patterns of electrical and/or magnetic flow through the scalp every millisecond. But the spatial resolution of EEG and MEG has been limited by the difficulty in finding the exact brain areas that generate the activity recorded outside the head. Mathematicians call this problem 'ill-posed' since an infinite number of current density source estimates are consistent with the data. However, almost none of these estimates are physiologically plausible. Thus, additional information about brain anatomy is needed to steer the solution toward a physiologically plausible answer. We use structural magnetic resonance (MR) head images to model the exact position and shape of the brain's outer rind or cortex, the scarf-sized, six-layer brain tissue in which all or nearly all the EEG and MEG signals are generated. Next, we make computer models of all possible dime-sized and smaller cortical patches that might generate some portion of the recorded data. Finally, we compute the maximally sparse solution to the inverse problem, i.e., the one that explains the recorded data with the lowest number of patches. Some of these may overlap to reveal larger source areas of any size and shape. We apply this 'sparse bayesian learning' (SBL) method to scalp maps of brain activity separated from the whole data by Independent Component Analysis (ICA), which finds distinct time signals that are mixed together in the scalp data. Combining ICA and SBL allows us to image unaveraged brain electromagnetic activity with high spatial and temporal resolution, allow imaging of coherent brain dynamic events processes accompanying behavior and experience."

Makeig, Ramirez and colleagues are preparing a series of manuscripts on these exciting new results.


For more information, contact:
Scott Makeig, Director and Research Scientist, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0961. Web site: http://sccn.ucsd.edu/~scott






Helpful Links
SCCN Overview
SCCN Home Page


Monday, October 14, 2024
About the Swartz Foundation...
 
The Swartz Foundation was established by Jerry Swartz (bio) in 1994 . . .
more>
 
Follow us...
 
The Swartz Foundation is on Twitter: SwartzCompNeuro
more>
 
 
About - Overview
 
 
Strategic Intent
 
 
Current Activities
 
 
Jerry Swartz - Biography
 
 
Books We Read
 
www.theswartzfoundation.org                           Copyright © The Swartz Foundation 2024