Disease State Prediction From Resting State Functional Connectivity

• Type of Material: Article
• Authors: R. Cameron Craddock, Georgia Institute of Technology; Paul Holtzheimer, Emory University; Xiaoping Hu, Emory University; Helen Mayberg, Emory University
• Language: English
• Date: 2009-12-01
• Publisher: Wiley
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © 2009 Wiley-Liss, Inc.
• Final Published Version (URL): http://dx.doi.org/10.1002/mrm.22159
• Title of Journal or Parent Work: Magnetic Resonance in Medicine
• ISSN: 0740-3194
• Volume: 62
• Issue: 6
• Start Page: 1619
• End Page: 1628
• Grant/Funding Information: This research was supported by NIMH P50 MH077083, 1R01MH073719 and Emory URC2004113.
• Keywords: FMRI DATA; DEPRESSION; HUMAN BRAIN; multivariate pattern analysis (MVPA); Life Sciences & Biomedicine; disease state prediction; support vector classification (SVC); major depressive disorder (MDD); CLASSIFICATION; COMPONENT ANALYSIS; SUPPORT VECTOR MACHINES; Science & Technology; functional connectivity; feature selection; Radiology, Nuclear Medicine & Medical Imaging; biomarker; PATTERN-ANALYSIS; SINGLE-SUBJECT; MRI; CINGULATE CORTEX
• Research Categories: Psychology, Behavioral; Engineering, Biomedical

Abstract

The application of multivoxel pattern analysis methods has attracted increasing attention, particularly for brain state prediction and real-time functional MRI applications. Support vector classification is the most popular of these techniques, owing to reports that it has better prediction accuracy and is less sensitive to noise. Support vector classification was applied to learn functional connectivity patterns that distinguish patients with depression from healthy volunteers. In addition, two feature selection algorithms were implemented (one filter method, one wrapper method) that incorporate reliability information into the feature selection process. These reliability feature selections methods were compared to two previously proposed feature selection methods. A support vector classifier was trained that reliably distinguishes healthy volunteers from clinically depressed patients. The reliability feature selection methods outperformed previously utilized methods. The proposed framework for applying support vector classification to functional connectivity data is applicable to other disease states beyond major depression.