Predicting MCI progression with FDG-PET and cognitive scores: a longitudinal study

BMC Neurology

Table 4 Comparison of the classification performance in the combination features

*Feature*	*ACC(%)*	*SEN(%)*	*SPE(%)*	*AUC*
Static	75.95	51.52	93.48	0.6614
Dynamic	87.38	87.88	86.96	0.8959
Cognitive	77.22	55.55	93.48	0.6414
Static & Dynamic	72.15	57.58	82.61	0.7444
Static & Cognitive	73.42	57.58	84.78	0.7345
Dynamic & Cognitive	87.34	75.76	95.65	0.9308
All	72.42	36.36	97.83	0.6443

ACC classification accuracy, SEN classification sensitivity, SPE classification specificity, AUC Area Under Curve. Static denotes the combination of all the static features, which is the same as Static_all in Table 2. Dynamic denotes the combination of all the dynamic features as shown in Table 3 (Dynamic_all). Cognitive denotes the combination of all the MMSE and ADAS-cog score features. All denotes the combination of all the types of features

ISSN: 1471-2377