What:Brain Imaging Series Lecture
When: Wednesday, January 12, 2011 3:15-4:15pm
Who: Free, Open to the public
Where: Center for Advanced Brain Imaging, Conference Room

Interdisciplinary Neuroimaging and Its Data Analysis

Michelle Wang

Department of Statistics and Department of Psychology
University of Illinois at Urbana-Champaign

Email: ymw@illinois.edu

Neuroimging is a highly interdisciplinary field that requires mathematical and statistical computation to solve complex neuroscience problems. The underlying computational issues are challenging and often hampered by the variability of brain anatomy and physiology, and the nature of the imaging data to be handled such as the presence of noise and correlation, and the sample and data sizes, etc. In this talk, after the brief introduction of neuroimaging, I will present the computational and statistical methods we have developed for several problems in the realms of brain morphometry, neural circuits and individual differences in learning from analyzing structural and functional magnetic resonance imaging (MRI) data. The discussion will include a description of the problem areas, an overview of the statistical techniques involved, and a presentation of results on simulated and real brain imaging data using these quantitative methods.

Chris Rorden and Paul Corballis will be teaching the CABI’s Image to inference course. This class is open to students and faculty, and can be taken for course credit (offered as PSYC 6042 at Georgia Tech) or audited free of charge. The course will describe how magnetic resonance imaging and electrophysiology can be used to understand brain function. Special emphasis is given to functional MRI (with SPM and FSL tutorials), though ERPs, VBM< DTI and other methods will be described. Lectures are from 9-12 Mondays thorughout the Georgia Tech term at the GSU/GT Center for Advanced Brain Imaging (see course web page for details and map).

Dear Colleagues:

I will be teaching IBS 575, Cerebral Cortex: Structures and Systems, during the Spring 2011 semester. I encourage you to make this course known to graduate students or highly qualified undergraduates who would like to take a comprehensive course on the organization of cerebral cortex.

I wish to make this course accessible for as many people as possible. Tentatively, I intend to teach it on Tuesdays from 5 – 8 pm at the Yerkes Primate Center. (Students should be aware that the Yerkes security gate is locked at 5 pm, so they will need to be prompt.) This time is not set in stone, however, and I will consider other times if there are serious conflicts with other courses.

Attached is a copy of the syllabus for the previous version of this course. It’s important to note, however, that there will be some changes in the 2011 offering. First, I will be teaching without a co-instructor. One result of this is that there will be less emphasis on functional imaging and more on evolution, development, and structural imaging (especially DTI and related techniques). This is not a methods course: the focus will be on results and interpretation.

All the best,

Mailing address
Todd M. Preuss, Ph.D.
Yerkes National Primate Research Center
Emory University
954 Gatewood Rd.
Atlanta, GA 30329 USA

404-727-8556 (office)
404-727-1331 (lab)
404-727-8070 (fax)


What:Brain Imaging Series Lecture
When: Wednesday, December 8, 2010 3:15-4:15pm
Who: Free, Open to the public
Where: Center for Advanced Brain Imaging, Conference Room

Applications of MVPA to the analysis of Resting State fMRI data:
Disease State Prediction, Brain State Prediction and Real-Time fMRI

Richard Cameron Craddock, Ph.D.
Postdoctoral Fellow
Virginia Tech Carilion Research Institute

Multi-voxel pattern analysis (MVPA) is sensitive to spatially
distributed patterns of coherent activation and is thus well suited to
the analysis of resting state fMRI data. An additional benefit of MVPA
is that the resulting trained model can be used to predict attributes
of never-before-seen resting state data. When applied at the group
level, MVPA identifies patterns of resting state functional
connectivity capable of distinguishing subject sub-populations (i.e.
disease state). At the individual level MVPA is capable of identifying
and prospectively measuring the activity of resting state networks.
This is particularly useful for performing brain-state prediction and
biofeedback fMRI experiments using these networks.

Short bio:
Cameron Craddock recently obtained his PhD in Electrical and Computer
Engineering from Georgia Tech under the supervision of Dr. Xiaoping Hu
and Dr. Helen Mayberg. He is currently a postdoctoral fellow at the
Virginia Tech Carilion Research Institute in Roanoke VA. His research
involves applying statistical signal processing and machine learning
analyses of functional neuroimaging techniques to identify biomarkers
of psychiatric disorders.