Bayesian Structure Learning for Functional Neuroimaging

Mijung Park, Oluwasanmi Koyejo, Joydeep Ghosh, Russell Poldrack, Jonathan Pillow
Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, PMLR 31:489-497, 2013.

Abstract

Predictive modeling of functional neuroimaging data has become an important tool for analyzing cognitive structures in the brain. Brain images are high-dimensional and exhibit large correlations, and imaging experiments provide a limited number of samples. Therefore, capturing the inherent statistical properties of the imaging data is critical for robust inference. Previous methods tackle this problem by exploiting either spatial sparsity or smoothness, which does not fully exploit the structure in the data. Here we develop a flexible, hierarchical model designed to simultaneously capture spatial block sparsity and smoothness in neuroimaging data. We exploit a function domain representation for the high-dimensional small-sample data and develop efficient inference, parameter estimation, and prediction procedures. Empirical results with simulated and real neuroimaging data suggest that simultaneously capturing the block sparsity and smoothness properties can significantly improve structure recovery and predictive modeling performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v31-park13a, title = {Bayesian Structure Learning for Functional Neuroimaging}, author = {Park, Mijung and Koyejo, Oluwasanmi and Ghosh, Joydeep and Poldrack, Russell and Pillow, Jonathan}, booktitle = {Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics}, pages = {489--497}, year = {2013}, editor = {Carvalho, Carlos M. and Ravikumar, Pradeep}, volume = {31}, series = {Proceedings of Machine Learning Research}, address = {Scottsdale, Arizona, USA}, month = {29 Apr--01 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v31/park13a.pdf}, url = {https://proceedings.mlr.press/v31/park13a.html}, abstract = {Predictive modeling of functional neuroimaging data has become an important tool for analyzing cognitive structures in the brain. Brain images are high-dimensional and exhibit large correlations, and imaging experiments provide a limited number of samples. Therefore, capturing the inherent statistical properties of the imaging data is critical for robust inference. Previous methods tackle this problem by exploiting either spatial sparsity or smoothness, which does not fully exploit the structure in the data. Here we develop a flexible, hierarchical model designed to simultaneously capture spatial block sparsity and smoothness in neuroimaging data. We exploit a function domain representation for the high-dimensional small-sample data and develop efficient inference, parameter estimation, and prediction procedures. Empirical results with simulated and real neuroimaging data suggest that simultaneously capturing the block sparsity and smoothness properties can significantly improve structure recovery and predictive modeling performance.} }
Endnote
%0 Conference Paper %T Bayesian Structure Learning for Functional Neuroimaging %A Mijung Park %A Oluwasanmi Koyejo %A Joydeep Ghosh %A Russell Poldrack %A Jonathan Pillow %B Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2013 %E Carlos M. Carvalho %E Pradeep Ravikumar %F pmlr-v31-park13a %I PMLR %P 489--497 %U https://proceedings.mlr.press/v31/park13a.html %V 31 %X Predictive modeling of functional neuroimaging data has become an important tool for analyzing cognitive structures in the brain. Brain images are high-dimensional and exhibit large correlations, and imaging experiments provide a limited number of samples. Therefore, capturing the inherent statistical properties of the imaging data is critical for robust inference. Previous methods tackle this problem by exploiting either spatial sparsity or smoothness, which does not fully exploit the structure in the data. Here we develop a flexible, hierarchical model designed to simultaneously capture spatial block sparsity and smoothness in neuroimaging data. We exploit a function domain representation for the high-dimensional small-sample data and develop efficient inference, parameter estimation, and prediction procedures. Empirical results with simulated and real neuroimaging data suggest that simultaneously capturing the block sparsity and smoothness properties can significantly improve structure recovery and predictive modeling performance.
RIS
TY - CPAPER TI - Bayesian Structure Learning for Functional Neuroimaging AU - Mijung Park AU - Oluwasanmi Koyejo AU - Joydeep Ghosh AU - Russell Poldrack AU - Jonathan Pillow BT - Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics DA - 2013/04/29 ED - Carlos M. Carvalho ED - Pradeep Ravikumar ID - pmlr-v31-park13a PB - PMLR DP - Proceedings of Machine Learning Research VL - 31 SP - 489 EP - 497 L1 - http://proceedings.mlr.press/v31/park13a.pdf UR - https://proceedings.mlr.press/v31/park13a.html AB - Predictive modeling of functional neuroimaging data has become an important tool for analyzing cognitive structures in the brain. Brain images are high-dimensional and exhibit large correlations, and imaging experiments provide a limited number of samples. Therefore, capturing the inherent statistical properties of the imaging data is critical for robust inference. Previous methods tackle this problem by exploiting either spatial sparsity or smoothness, which does not fully exploit the structure in the data. Here we develop a flexible, hierarchical model designed to simultaneously capture spatial block sparsity and smoothness in neuroimaging data. We exploit a function domain representation for the high-dimensional small-sample data and develop efficient inference, parameter estimation, and prediction procedures. Empirical results with simulated and real neuroimaging data suggest that simultaneously capturing the block sparsity and smoothness properties can significantly improve structure recovery and predictive modeling performance. ER -
APA
Park, M., Koyejo, O., Ghosh, J., Poldrack, R. & Pillow, J.. (2013). Bayesian Structure Learning for Functional Neuroimaging. Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 31:489-497 Available from https://proceedings.mlr.press/v31/park13a.html.

Related Material