2006 PBAIC Results
The 2006 Pittsburgh Brain Activity Interpretation Competition was a very successful competition involving prediction of the subjective experience of movie viewing from fMRI data for prize money. The competition generated widespread interest as evidenced by 25,000 website hits and 273 groups in 31 countries downloading fMRI data sets. It was judged by a board of experts from the field of neuroimaging and included an objective scoring method. The competition attracted coverage in the popular and scientific press and was reported to have advanced the field and held up as a model for scientific competitions (see What's on your mind? Nature Neuroscience 2006, Vol. 8, p. 981). Competitors achieved impressive prediction accuracy and brought new talent and methods into the field.
Please visit the 2006 PBAIC web site for more information.
\
Diego Sona, Greg Siegle, Emanuele Olivett, Sriharsha Veeramachaneni,
Alexis Battle, Greg Stephens, Walter Schneider
Top 3 Winners
1. Emanuele Olivetti, Diego Sona, Sriharsha Veeramachaneni
Gaussian process regression and recurrent neural networks for fMRI image classification
ITC-IRST; Italy
Summary Page| Methods Description | Score Sheet | Poster | Slides
$10,000 First Place (US Dollars)
2. Denis Chigirev, Greg Stephens and The Princeton EBC team
Predicting Base Features with Supervoxels
Princeton University; United States of America
Summary Page | Methods Description | Score Sheet | Slides
$5,000 Second Place (US Dollars)
3. Alexis Battle, Gal Chechik, Daphne Koller
Temporal and Cross-Subject Probabilistic Models for fMRI Prediction Tasks
Stanford University; United States of America
Summary Page | Methods Description | Score Sheet | Slides
$2,000 Third Place (US Dollars)
Honorable Mentions
Stephen M. LaConte
Brain Reading with Selective Training Data, Transduction, and Behavior Based Group Analysis
Emory University; United States of America
Summary Page| Methods Description | Score Sheet
*Best Prediction of All Categories
S. Ghebreab, P. Adriaans, A.W.M. Smeulders
Incremental ClusterWise Regression Analysis of Functional Stimulation-Activation Data
University of Amsterdam; Netherlands
Summary Page | Methods Description| Score Sheet | Poster
*Best Prediction of Actor Exemplars
Paul F. Rodriguez
Fast Prediction Using Large Numbers of Voxels with Kernel Based Non-linear Partial Least Squares Regression
University of California, Irvine; United States of America
Summary Page | Methods Description | Score Sheet
*Best Prediction of Internal Assessment
