The Kernel Two-Sample Test for Brain Networks
This work addresses the need for efficient and accurate statistical testing in clinical and neuroscientific studies of brain networks, though it is incremental as it adapts an existing test to a specific domain.
The authors tackled the problem of detecting differences between two populations of brain networks, proposing the kernel two-sample test (KTST) as a direct method that avoids intermediate classifiers, and found it provides similar results with less computation and lower Type II error in low sample sizes.
In clinical and neuroscientific studies, systematic differences between two populations of brain networks are investigated in order to characterize mental diseases or processes. Those networks are usually represented as graphs built from neuroimaging data and studied by means of graph analysis methods. The typical machine learning approach to study these brain graphs creates a classifier and tests its ability to discriminate the two populations. In contrast to this approach, in this work we propose to directly test whether two populations of graphs are different or not, by using the kernel two-sample test (KTST), without creating the intermediate classifier. We claim that, in general, the two approaches provides similar results and that the KTST requires much less computation. Additionally, in the regime of low sample size, we claim that the KTST has lower frequency of Type II error than the classification approach. Besides providing algorithmic considerations to support these claims, we show strong evidence through experiments and one simulation.