Robust feature selection in resting-state fMRI connectivity based on population studies
We propose an alternative to univariate statistics for identifying population differences in functional connectivity. Our feature selection method is based on a procedure that searches across subsets of the data to isolate a set of robust, predictive functional connections. The metric, known as the...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2012
|
| Online Access: | http://hdl.handle.net/1721.1/72387 https://orcid.org/0000-0003-2516-731X https://orcid.org/0000-0002-2683-5888 |
| _version_ | 1826204206556512256 |
|---|---|
| author | Venkataraman, Archana Kubicki, Marek Westin, Carl-Fredrik Golland, Polina |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Venkataraman, Archana Kubicki, Marek Westin, Carl-Fredrik Golland, Polina |
| author_sort | Venkataraman, Archana |
| collection | MIT |
| description | We propose an alternative to univariate statistics for identifying population differences in functional connectivity. Our feature selection method is based on a procedure that searches across subsets of the data to isolate a set of robust, predictive functional connections. The metric, known as the Gini Importance, also summarizes multivariate patterns of interaction, which cannot be captured by univariate techniques. We compare the Gini Importance with univariate statistical tests to evaluate functional connectivity changes induced by schizophrenia. Our empirical results indicate that univariate features vary dramatically across subsets of the data and have little classification power. In contrast, relevant features based on Gini Importance are considerably more stable and allow us to accurately predict the diagnosis of a test subject. |
| first_indexed | 2024-09-23T12:50:34Z |
| format | Article |
| id | mit-1721.1/72387 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T12:50:34Z |
| publishDate | 2012 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/723872022-10-01T11:26:14Z Robust feature selection in resting-state fMRI connectivity based on population studies Venkataraman, Archana Kubicki, Marek Westin, Carl-Fredrik Golland, Polina Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Golland, Polina Venkataraman, Archana Westin, Carl-Fredrik Golland, Polina We propose an alternative to univariate statistics for identifying population differences in functional connectivity. Our feature selection method is based on a procedure that searches across subsets of the data to isolate a set of robust, predictive functional connections. The metric, known as the Gini Importance, also summarizes multivariate patterns of interaction, which cannot be captured by univariate techniques. We compare the Gini Importance with univariate statistical tests to evaluate functional connectivity changes induced by schizophrenia. Our empirical results indicate that univariate features vary dramatically across subsets of the data and have little classification power. In contrast, relevant features based on Gini Importance are considerably more stable and allow us to accurately predict the diagnosis of a test subject. National Science Foundation (U.S.). Career Award (0642971) National Institutes of Health (U.S.) (R01MH074794) National Defense Science and Engineering Graduate Fellowship (Department of Defense Fellowship) National Institutes of Health (U.S.) (National Alliance for Medical Image Analysis) (NIH NIBIB NAMIC U54-EB005149), National Institutes of Health (U.S.) (Neuroimaging Analysis Center) (NIH NCRR NAC P41-RR13218) National Science Foundation (U.S.). Career Award (0642971) 2012-08-28T19:24:52Z 2012-08-28T19:24:52Z 2010-06 2010-06 Article http://purl.org/eprint/type/ConferencePaper 978-1-4244-7029-7 http://hdl.handle.net/1721.1/72387 Venkataraman, Archana et al. “Robust Feature Selection in Resting-state fMRI Connectivity Based on Population Studies.” IEEE, 2010. 63–70. © Copyright 2010 IEEE https://orcid.org/0000-0003-2516-731X https://orcid.org/0000-0002-2683-5888 en_US http://dx.doi.org/10.1109/CVPRW.2010.5543446 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Institute of Electrical and Electronics Engineers (IEEE) IEEE |
| spellingShingle | Venkataraman, Archana Kubicki, Marek Westin, Carl-Fredrik Golland, Polina Robust feature selection in resting-state fMRI connectivity based on population studies |
| title | Robust feature selection in resting-state fMRI connectivity based on population studies |
| title_full | Robust feature selection in resting-state fMRI connectivity based on population studies |
| title_fullStr | Robust feature selection in resting-state fMRI connectivity based on population studies |
| title_full_unstemmed | Robust feature selection in resting-state fMRI connectivity based on population studies |
| title_short | Robust feature selection in resting-state fMRI connectivity based on population studies |
| title_sort | robust feature selection in resting state fmri connectivity based on population studies |
| url | http://hdl.handle.net/1721.1/72387 https://orcid.org/0000-0003-2516-731X https://orcid.org/0000-0002-2683-5888 |
| work_keys_str_mv | AT venkataramanarchana robustfeatureselectioninrestingstatefmriconnectivitybasedonpopulationstudies AT kubickimarek robustfeatureselectioninrestingstatefmriconnectivitybasedonpopulationstudies AT westincarlfredrik robustfeatureselectioninrestingstatefmriconnectivitybasedonpopulationstudies AT gollandpolina robustfeatureselectioninrestingstatefmriconnectivitybasedonpopulationstudies |