Hybrid ICA-Seed-Based Methods for fMRI Functional Connectivity Assessment: A Feasibility Study
Figure 7
ROC curves for Subject 1, on the interval from 0 to 0.05. Subject 1’s task-related, GLM-derived spatial map was used as the “true” standard of comparison. The points where the number of false positives (FPs) was equal to the number of false negatives (FNs) fall along the FP = FN line shown above. This helps to explain why the results of the ROC PAUC analyses (Experiments Criterion 5) were so similar to the results of the threshold-adjusted overlap analyses (Experiments Criterion 4).
