Table 3.
Student's t test P-values are shown for comparing the SVD and the HO-SVD feature sets to the Raw feature set for each of the diagnostic classification tasks.
| SVD > Raw | HO-SVD > Raw | HO-SVD > SVD | ||
|---|---|---|---|---|
| AD vs. NC | Accuracy | 0.6393 | 0.4718 | 0.3494 |
| Sensitivity | 0.5000 | 0.2456 | 0.2499 | |
| Specificity | 0.6734 | 0.6363 | 0.4686 | |
| AUC | 0.4101 | 0.2963 | 0.3786 | |
| AD vs. MCI | Accuracy | 0.5805 | 0.0005 | 0.0003 |
| Sensitivity | 0.7440 | 0.2790 | 0.1017 | |
| Specificity | 0.5165 | 0.0010 | 0.0009 | |
| AUC | 0.2860 | 0.0267 | 0.0554 | |
| MCI vs. NC | Accuracy | 0.3473 | 0.0082 | 0.0102 |
| Sensitivity | 0.3562 | 0.0125 | 0.0382 | |
| Specificity | 0.5000 | 0.3553 | 0.3425 | |
| AUC | 0.4051 | 0.0286 | 0.0474 |
A Bonferroni correction was adopted here, to account for multiple testing. As there are four measures including Accuracy, Sensitivity, Specificity, and AUC, the corrected P threshold in each column is 0.05/4 = 0.0125. P < 0.0125 are marked in red.