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 5
10-fold cross-validated classification performance for HC vs. MCI.
| Features measure | ELM | SVM-RBF | SVM-linear | ACC% | SEN% | SPE% | AUC | ACC% | SEN% | SPE% | ACC% | SEN% | SPE% |
| Cortical thickness | 75.40 | 84.07 | 67.90 | 0.83 | 74.17 | 80.81 | 71.23 | 67.34 | 73.17 | 77.93 | Surface area | 74.15 | 65.20 | 83.85 | 0.84 | 63.54 | 67.67 | 66.47 | 68.72 | 74.43 | 67.35 | Volume | 77.84 | 88.41 | 72.80 | 0.84 | 65.47 | 74.73 | 68.75 | 72.83 | 78.27 | 65.70 | CSF | 83.30 | 73.93 | 84.02 | 0.85 | 73.39 | 75.25 | 66.83 | 75.80 | 73.38 | 78.14 | APOE + MMSE | 79.37 | 93.25 | 72.97 | 0.89 | 68.29 | 81.73 | 76.68 | 70.23 | 81.24 | 78.05 | Concatenation (all_features_set) | 91.72 | 91.23 | 99.13 | 0.93 | 81.70 | 83.68 | 87.49 | 85.65 | 78.54 | 89.26 |
|
|