Fig 2. Conceptual description of the problem designed to test the sensitivity of AMURA to micro-structural changes.
(A) The pixels in the image are split into 2 classes by thresholding the corresponding MAPL measure (RTOP in the example). (B) Each one of the methods to be tested: MAPL, MAP-MRI, RBF, or AMURA (AMURA in the example) is used to compute this same measure, and a ROC curve is calculated with the classes defined in (A) as the target.