Denouements of machine learning and multimodal diagnostic classification of Alzheimer’s disease

Visual Computing for Industry, Biomedicine, and Art

Table 4 Classification accuracy for individual and combined image modalities

	T-1 MRI	FDG-PET		rCBF-SPECT (GN)		rCBF-SPECT (CN)		T-1 MRI + FDG-PET		T-1 MRI + rCBF-SPECT (GN)		T-1 MRI + rCBF-SPECT (CN)
	T-1 MRI	(GN)	(CN)	PVE	No PVE	PVE	No PVE	(GN)	(CN)	PVE	No PVE	PVE	No PVE
AUC	0.67	0.81	0.77	0.75	0.76	0.75	0.79	0.81	0.82	0.78	0.78	0.74	0.76
TP (%)	50.00	70.00	70.00	60.00	70.00	60.00	70.00	75.00	70.0	75.0	80.00	70.00	75.00
TN (%)	66.67	66.67	72.22	77.78	66.67	83.33	77.78	66.67	72.22	72.22	66.67	66.67	72.22
BA (%)	58.33	68.33	71.11	68.89	68.33	71.67	73.89	70.83	71.11	73.61	73.33	68.33	73.61
p-value*	0.1379	0.0242	0.0157	0.0210	0.0214	0.0071	0.0043	0.0198	0.0014	0.0008	0.0011	0.0253	0.0003

BA Balanced accuracy, TN True negative, TP True positive, CN Cerebellar normalization, GN Global normalization, PVE Partial volume effect correction
*Nonparametric statistical significance