Research Article
An Efficient Combination among sMRI, CSF, Cognitive Score, and APOE ε4 Biomarkers for Classification of AD and MCI Using Extreme Learning Machine
Table 3
10-fold cross-validated classification performance for AD vs. HC.
| Features measure | ELM | SVM-RBF | SVM-linear | ACC% | SEN% | SPE% | AUC | ACC% | SEN% | SPE% | ACC% | SEN% | SPE% |
| Cortical thickness | 85.13 | 92.07 | 85.44 | 0.86 | 80.68 | 85.68 | 83.81 | 77.17 | 83.17 | 73.81 | Surface area | 82.30 | 89.30 | 91.54 | 0.85 | 76.16 | 86.16 | 75.14 | 78.67 | 87.67 | 72.62 | Volume | 87.90 | 93.87 | 84.75 | 0.88 | 79.29 | 74.29 | 78.33 | 80.00 | 77.10 | 63.24 | CSF | 89.73 | 96.33 | 87.38 | 0.89 | 79.05 | 89.05 | 74.17 | 75.33 | 83.33 | 68.57 | APOE + MMSE | 86.45 | 95.13 | 87.00 | 0.93 | 82.03 | 87.03 | 84.67 | 84.16 | 83.05 | 78.79 | Concatenation (all_features_set) | 97.31 | 98.04 | 96.28 | 0.97 | 91.33 | 93.33 | 87.57 | 93.50 | 95.5 | 90.58 |
|
|