Table 4.
Characteristics of the ROC curve for (A) the full data set without MRI and (B) ROC curve characteristics for the subset with MRI imaging data
| (A) ROC characteristics without MRI | |||||||
|---|---|---|---|---|---|---|---|
| Classification model | Sensitivity cut-off % | SN % | SP % | PPV % | NPV % | ACC % | ROC (AUC) |
| Protein + APOE | 30 | 30.8 | 92.9 | 57.1 | 81.3 | 87.2 | 0.84 |
| Protein only | 30 | 30.8 | 92.9 | 57.1 | 81.3 | 87.2 | 0.78 |
| Protein + APOE | 50 | 53.9 | 88.1 | 58.3 | 86.1 | 80.0 | 0.84 |
| Protein only | 50 | 43.8 | 84.6 | 53.9 | 78.6 | 72.7 | 0.78 |
| Protein + APOE | 85 | 84.6 | 88.1 | 68.8 | 94.9 | 87.2 | 0.84 |
| Protein only | 85 | 84.6 | 71.4 | 47.8 | 93.8 | 74.5 | 0.78 |
| (B) ROC characteristics with MRI | |||||||
|---|---|---|---|---|---|---|---|
| Classification model | Sensitivity cut-off % | SN % | SP % | PPV % | NPV % | ACC % | ROC (AUC) |
| Protein + APOE + MRI | 30 | 33.3 | 96.7 | 66.7 | 87.9 | 86.1 | 0.75 |
| Protein only | 30 | 33.3 | 93.3 | 50.0 | 87.5 | 83.3 | 0.82 |
| MRI only | 30 | 33.3 | 80.0 | 25.0 | 85.7 | 72.2 | 0.54 |
| Protein + APOE + MRI | 50 | 50.0 | 90.0 | 50.0 | 90.0 | 83.3 | 0.75 |
| Protein only | 50 | 50.0 | 86.7 | 42.9 | 89.7 | 80.6 | 0.82 |
| MRI only | 50 | 50.0 | 63.3 | 21.3 | 86.4 | 61.1 | 0.54 |
| Protein + APOE + MRI | 85 | 83.3 | 60.0 | 29.4 | 94.7 | 63.9 | 0.75 |
| Protein only | 85 | 83.3 | 66.7 | 33.3 | 95.2 | 69.4 | 0.82 |
| MRI only | 85 | 83.3 | 13.3 | 16.1 | 80.0 | 25.0 | 0.54 |
Abbreviations: ROC, receiver operating characteristic; MRI, magnetic resonance imaging; SN, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value; ACC, accuracy; AUC, ROC area under curve for the protein and APOE classifiers.
NOTE. Three different sensitivity cut-off points were investigated.