Unsupervised Spatiotemporal Analysis of FMRI Data Using Graph-Based Visualizations of Self-Organizing Map
Santosh B. Katwal, John C. Gore, René Marois, and Baxter P. Rogers
Functional magnetic resonance imaging (fMRI) data are commonly analyzed voxel-by-voxel using linear regression models (statistical parametric mapping) which requires information about stimulus timing and assumptions about the shape and timing of the hemodynamic response. This approach may be too restrictive to capture the broad range of possible brain activation patterns in space and time and across subjects. We present a multivariate data-driven approach using self-organizing maps that overcome the aforementioned limitations. We propose novel graph-based visualizations of self-organizing maps for unsupervised fMRI analysis. A self-organizing map is an artificial neural network model that transforms high-dimensional data into a low-dimensional (often a 2-D) map using unsupervised learning. However, a postprocessing scheme is necessary to correctly interpret similarity between neighboring node prototypes (feature vectors) on the output map and delineate clusters and features of interest in the data. In this paper, we used graph-based visualizations to capture fMRI data features based upon 1) the distribution of data across the receptive fields of the prototypes (density-based connectivity), and 2) temporal similarities (correlations) between the prototypes (correlation-based connectivity). We applied this approach to identify task-related brain areas in an fMRI reaction time experiment involving a visuo-manual response task, and we correlated the time-to-peak of the fMRI responses in these areas with reaction time. Visualization of self-organizing maps outperformed independent component analysis and voxelwise univariate linear regression analysis in identifying and classifying relevant brain regions. This was verified by observing a statistically strong linear relationship between induced and measured timing differences. We conclude that the graph-based visualizations of self-organizing maps help in advanced visualization of cluster boundaries in fMRI data enabling the separation of regions with small differences in the timings of their brain responses.