Table 3.

P values of the non-parametric permutation tests to test statistical differences between classifiers based on a) mean area under the ROC curve (AUC) and b) mean accuracy

	CBF vs. VBM-GM	FA vs. VBM-WM	GM combination vs. VBM-GM	WM combination vs. VBM-WM	Full combination vs. VBM-Brain
a) Mean area under the ROC curve (AUC)
AD-CN	0.810	0.834	0.798	0.452	0.552
FTD-CN	0.388	0.466	0.818	0.816	0.818
AD-FTD	0.752	0.668	0.472	0.322	0.052*
AD-FTD-CN	0.664	0.892	0.546	0.220	0.028*
b) Mean accuracy
AD-CN	0.476	0.688	0.118	0.222	0.540
FTD-CN	0.210	0.324	0.624	0.462	0.998
AD-FTD	0.476	0.980	0.224	0.898	0.122
AD-FTD-CN	0.566	0.920	0.340	0.176	0.050*

^* Significant difference (p ≤ 0.05)

AD Alzheimer’s disease, AUC area under the receiver-operating characteristic curve, CBF cerebral blood flow, CN cognitively normal controls, FA fractional anisotropy, FTD frontotemporal dementia, GM grey matter, ROC receiver-operating characteristic, VBM voxel-based morphometry, WM white matter