Table 3.
Method | Region | SUVReHC | SUVRAD | SUVRcutoff | Sensitivity | Specificity | AUC |
---|---|---|---|---|---|---|---|
MRI-based ROIs | Frontal | 1.15 ± 0.07 | 1.57 ± 0.19 | 1.31 | 95% | 97% | 0.98 |
Lateral temporal | 1.15 ± 0.05 | 1.51 ± 0.17 | 1.26 | 97% | 98% | 0.96 | |
Occipital | 1.20 ± 0.06 | 1.43 ± 0.16 | 1.29 | 88% | 95% | 0.94 | |
Parietal | 1.13 ± 0.08 | 1.51 ± 0.16 | 1.26 | 97% | 98% | 0.98 | |
Anterior cingulate | 1.28 ± 0.09 | 1.70 ± 0.22 | 1.43 | 92% | 97% | 0.98 | |
Posterior cingulate | 1.33 ± 0.09 | 1.77 ± 0.21 | 1.47 | 93% | 97% | 0.97 | |
Precuneus | 1.15 ± 0.08 | 1.60 ± 0.19 | 1.28 | 95% | 98% | 0.98 | |
Composite | 1.21 ± 0.06 | 1.58 ± 0.17 | 1.38 | 93% | 100% | 0.98 | |
CL ROIs | Cortex |
1.05 ± 0.06 6.8 ± 8.8 CL |
1.54 ± 0.22 81.0 ± 33.2 CL |
1.24 35.7 CL |
95% | 100% | 0.98 |
eHC elderly healthy controls, Aβ amyloid-beta, SUVReHC SUVR (mean ± SD) of the elderly healthy controls, SUVRAD SUVR (mean ± SD) of the Alzheimer’s disease patients, AUC area under the receiver operating curve, SUVRcutoff SUVR cutoff obtained from the ROC analysis using visual assessment as standard of truth, MRI magnetic resonance imaging, ROI region of interest, CL centiloid