Signficance determination for the scale-space primal sketch by comparison of statistics of scale-space blob volumes computed from PET signals vs. residual noise

摘要

A dominant approach to brain mapping is to define functional regions in the brain by analyzing brain activation images obtained by PET or fMRI. In [1], it has been shown that the scale-space primal sketch provides a useful tool for such analysis. Some attractive properties of this method are that it only makes few assumptions about the data and the process for extracting activations is fully automatic. In the present version of the scale-space primal sketch, however, there is no method for determining p-values. The purpose here is to present a new methodology for addressing this question, by introducing a descriptor referred to as the -curve, which serves as a first step towards determining the probability of false positives, i.e. alpha.