New Directions in
Statistical Learning for Meaningful and Reproducible fMRI Analysis
NIPS 08 workshops, December 13, Whistler, Canada
Organizers
Melissa Carroll is a graduate student in Computer Science and Neuroscience at Princeton University, advised by Robert Schapire. She earned a BA in Psychology from Binghamton University, an MS in Computer Science from Pace University, and an MA in Computer Science from Princeton University. From 1999-2004, she worked as a Research Data Specialist in the Psychiatry department at Weill Medical College of Cornell University, analyzing clinical and neuroimaging data. Since 2004, her graduate work has focused on the use of statistical learning techniques for neuroscientific discovery from fMRI data. In 2007 and 2008, she interned with the Biometaphorical Computing Group at the IBM TJ Watson Research Center, focusing on analysis of fMRI data.
Relevant publications:
- Carroll, M. K., Cecchi, G., Rish, I., Garg, R., Rao, A. R. (2008) Prediction and Interpretation of Distributed Neural Activity with Sparse Models. Neuroimage (under review).
- Carroll, M. K., Cecchi, G., Rish, I., Garg, R., Rao, A. R. (2008) Beyond Prediction: Discovering Distributed Patterns of Brain Activity from fMRI Data Via Sparse Regression. Abstract to be presented at the Society for Neuroscience meeting, Washington, DC.
- Carroll, M.K., Dudik, M. (2007) Feature Induction on fMRI Images Using Regularized Logistic Regression. Abstract presented at the 13th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, IL.
- Carroll, M.K., Dudik, M., Schapire, R.E., Norman, K.A. (2006) Feature Induction Using Boosting and Logistic Regression on fMRI Images. Presentation at the NIPS 2006 Workshop on New Directions on Decoding Mental States from fMRI Data. Whistler, BC.
- Carroll, M.K., Norman, K.A., Haxby, J.V., Schapire, R.E. (2006) Exploiting Spatial Information to Improve fMRI Pattern Classification. Abstract presented at the 12th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Florence, Italy.
Irina Rish is a research staff member at the Biometaphorical Computing Department which is a part of the Computational Biology Center at the IBM T. J. Watson Research Center. She received an M.S. in applied mathematics from Moscow Gubkin Institute, Russia, and a Ph.D. in computer science from the University of California, Irvine. Dr. Rish's primary research interests are in the areas of probabilistic inference, machine learning, and information theory, and their applications to large-scale data analysis problems in biology and neuroscience. Her current research focus is on applyng machine-learning techniques to neuroscience, and particularly on statistical analysis of fMRI data using sparse regression and dimensionality reduction. In the past, she has also worked on efficient approximations of probabilistic inference in Bayesian networks, active learning, collaborative prediction, sparse regression and sparse matrix factorization, and their applications to autonomic computing (self-managing computer systems). She over 40 conference and journal publications on the above topics. Dr. Rish taught several machine learning courses at the Electrical Engineering and Computer Science departments of Columbia University as an adjunct professor, and co-organized several machine-learning workshops, including ICML workshop in 2000, NIPS workshops in 2003, 2005 and 2006, ECML workshop in 2006, and, most recently, ICML 2008 workshop on Sparse Regression and Variable Selection.
Relevant publications:
- Rish, I., Grabarnik, G., Cecchi, G., Pereira, F., Gordon., G. (2008) Closed-form Supervised Dimensionality Reduction with Generalized Linear Models, in Proceedings of ICML 2008, Helsinki, Finland.
- G. Cecchi, I. Rish, R. Rao, R. Garg. (2008) Prediction of Brain Activity based on Elastic Net Algorithm.Abstract in PBAIC workshop at the 13th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, IL.
- Carroll, M. K., Cecchi, G., Rish, I., Garg, R., Rao, A. R. (2008) Prediction and Interpretation of Distributed Neural Activity with Sparse Models. Neuroimage (under review).
- Thyreau, B., Garg, R., Cecchi, G.A., Plaze, M., Rao, A.R., Rish, I. Martinot, J-L., Poline, J-B. (2008) Schizophrenia as a Disruption of Functional Connectivity Patterns. Abstract presented at the 14th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Florence, Italy.
- Garg, R., Cecchi, G., Rao, A.R., Rish, I. (2008) A comparison of fMRI activation maps obtained using GLM with maps generated using network-based analysis techniques. Abstract to be presented at Society for Neuroscience meeting. Washington, DC.
Francisco Pereira is a post-doctoral researcher at the Princeton Neuroscience Institute, working with Matthew Botvinick, Ken Norman and David Blei. He was a graduate student under Tom Mitchell and Geoff Gordon at Carnegie Mellon University, and graduated with a PhD in Computer Science/Neural Basis of Cognition in December 2007. His broad area of research is the development and adaptation of machine learning methods for the purpose of answering scientific questions and facilitating discovery in cognitive neuroscience. He was the lead organizer on the 2006 NIPS Workshop on New Directions on Decoding Mental States from fMRI Data.
Relevant publications:
- Pereira F., Mitchell T., Botvinick M. Machine learning classifiers and fMRI: a tutorial overview (under review)
- Pereira, F. (2007) Beyond Brain Blobs: Machine Learning Classifiers as Instruments for Analyzing Functional Magnetic Resonance Imaging Data. Ph.D. dissertation, Carnegie Mellon University
- Pereira, F., Gordon, G. (2006) The Support Vector Decomposition Machine. International Conference on Machine Learning.
- Chen X., Pereira F., Lee W., Strother S., Mitchell T. (2006) Exploring predictive and reproducible modeling with the single-subject FIAC data set. Hum Brain Mapping 27(5):452-61
- Mitchell T., Hutchinson R., Niculescu S., Pereira F., Wang X., Just M., Newman S. (2004) Learning to Decode Cognitive States from Brain Images. Machine Learning Journal, Vol. 57, Issue 1-2, pp. 145-175.
Guillermo Cecchi received an education in Physics (MSc, University of La Plata, Argentina, 1991), Physics and Biology (PhD, The Rockefeller University, 1994-1999), and Imaging in Psychiatry (Postdoctoral Fellow, Cornell University 2000-2001). In 2001 he joined IBM Research to be part of the Biometaphorical Computing project, where he has been working on computational approaches to brain function and systems biology. His research interests have covered diverse aspects of theoretical biology, including Brownian transport, molecular computation, spike reliability in neurons, song production and representation in songbirds, statistics of natural images and visual perception, statistics of natural language, and brain imaging. Recently, he has pioneered the use of statistical network theory for the analysis and modeling of functional brain networks.
Relevant publications:
- Rish, I., Grabarnik, G., Cecchi, G., Pereira, F., Gordon., G. (2008) Closed-form Supervised Dimensionality Reduction with Generalized Linear Models, in Proceedings of ICML 2008, Helsinki, Finland.
- Carroll, M. K., Cecchi, G., Rish, I., Garg, R., Rao, A. R. (2008) Prediction and Interpretation of Distributed Neural Activity with Sparse Models. Neuroimage (under review).
- Cecchi, G.A., Garg, R., Rao, A.R. (2008) Inferring brain dynamics using Granger causality on fMRI data. Proceedings of the IEEE International Symposium on Biomedical Imaging, Paris.
- Cecchi, G.A., Rao, A.R., Centeno, M.V., Baliki, M., Apkarian, A.V., Chialvo, D.R. (2007) Identifying directed links in large scale functional networks: application to brain fMRI. BMC Cell Biology 8(Suppl 1):S5.
- Eguiluz, V.M., Chialvo, D.R., Cecchi, G.A., Baliki, M., Apkarian, A.V. (2005) Scale-free functional brain networks. Physical Review Letters 94, 018102.
- Eguiluz, V.M., Sosa, Y. Baliki, M., Cecchi, G.A., Chialvo, D.R., Apkarian, A.V. (2003) Analysis of brain activity as a massively interconnected dynamical network with fMRI. Abstract presented at the Human Brain Mapping Meeting.