Skip to main content
. 2019 May 31;13:509. doi: 10.3389/fnins.2019.00509

Table 5.

Summary of the models trained from single modality protocols and three multi-modality protocols for CN vs. AD task.

Method ACC SEN SPE AUC
MRI 81.19% 79.27% 83.45% 83.67%
PET 89.11% 90.24% 87.77% 92.69%
A 87.79% 85.98% 89.93% 89.42%
B1 90.10% 90.85% 89.21% 90.84%
B2 89.44% 89.02% 89.93% 92.01%

The Paired dataset was used. The best results were indicated in bold.