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

BMC Neurology

Table 3 Comparison of the classification performance in dynamic features

*Feature*	*ACC(%)*	*SEN(%)*	*SPE(%)*	*AUC*
Dynamic_1	88.61	81.82	93.48	0.9351
Dynamic_2	77.21	75.76	78.26	0.8063
Dynamic_3	82.28	73.91	93.94	0.9289
R1	60.76	6.32	100	0.8524
R2	64.56	15.22	100	0.6278
R3	64.56	54.55	71.74	0.5738
Dynamic_all	87.38	87.88	86.96	0.8959

ACC classification accuracy, SEN classification sensitivity, SPE classification specificity. AUC Area Under Curve, Dynamic_1 dynamic feature calculated with Static_mbl and Static_m6 in Table 2, Dynamic_2 dynamic feature calculated with Static_mbl and Static_m12, Dynamic_3 dynamic feature calculated with Static_mbl and Static_m18, R1 metabolic change rate in the 6th month after baseline, R2 metabolic change rate in the 12th month after baseline, R3 metabolic change rate in the 18th month after baseline, Dynamic_all combining all the dynamic features

ISSN: 1471-2377