Research Article
Automatic Diagnosis of Alzheimer’s Disease and Mild Cognitive Impairment Based on CNN + SVM Networks with End-to-End Training
Table 6
Evaluation of the proposed method applied to 3-category classification in terms of ACC (%).
| Method | Training set | Testing set | AD | MCI | NC | Average | AD | MCI | NC | Average |
| Cabral et al. [9] | — | — | — | — | — | — | —- | 66.78 | 3DCNN | 99.85 | 98.62 | 99.89 | 99.45 | 65.63 | 62.12 | 70.43 | 65.66 | CNN + BGRU [14] | 97.75 | 99.89 | 99.86 | 99.17 | 58.65 | 66.22 | 68.28 | 65.53 | ADCNN [16] | 99.81 | 98.35 | 99.99 | 99.38 | 65.16 | 63.25 | 68.63 | 65.44 | Proposed | 99.17 | 97.83 | 99.37 | 98.79 | 73.42 | 67.86 | 72.28 | 71.19 |
|
|