A supervised clustering approach for fMRI-based inference of brain states

• Published in: Pattern recognition Vol. 45; no. 6; pp. 2041 - 2049
• Main Authors: Michel, Vincent, Gramfort, Alexandre, Varoquaux, Gaël, Eger, Evelyn, Keribin, Christine, Thirion, Bertrand
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.06.2012; Elsevier
• Subjects: Brain; Brain reading; Clustering; Computer Science; Computer Vision and Pattern Recognition; Dimension reduction; Feature agglomeration; fMRI; Hierarchical clustering; Inference; Multi-scale analysis; Prediction; Reduction; Regression; Trees; fMRI; Dimension reduction; Multi-scale analysis; Prediction; Feature agglomeration; Hierarchical clustering; Brain reading; multi-scale analysis; prediction; hierarchical clustering; feature agglomeration; brain reading; dimension reduction
• ISSN: 0031-3203; 1873-5142
• DOI: 10.1016/j.patcog.2011.04.006

We propose a method that combines signals from many brain regions observed in functional Magnetic Resonance Imaging (fMRI) to predict the subject's behavior during a scanning session. Such predictions suffer from the huge number of brain regions sampled on the voxel grid of standard fMRI data sets: the curse of dimensionality. Dimensionality reduction is thus needed, but it is often performed using a univariate feature selection procedure, that handles neither the spatial structure of the images, nor the multivariate nature of the signal. By introducing a hierarchical clustering of the brain volume that incorporates connectivity constraints, we reduce the span of the possible spatial configurations to a single tree of nested regions tailored to the signal. We then prune the tree in a supervised setting, hence the name supervised clustering, in order to extract a parcellation (division of the volume) such that parcel-based signal averages best predict the target information. Dimensionality reduction is thus achieved by feature agglomeration, and the constructed features now provide a multi-scale representation of the signal. Comparisons with reference methods on both simulated and real data show that our approach yields higher prediction accuracy than standard voxel-based approaches. Moreover, the method infers an explicit weighting of the regions involved in the regression or classification task. ► We provide and efficient dimension reduction method for structured data. ► It is based on a feature agglomeration idea, and builds a multi-scale model of the data. ► We develop the framework for prediction and classification settings. ► Our approach outperforms standard techniques for activation image-based prediction of brain states. ► It provides an interpretable representation of informative features in image-based prediction problems.