Table 3.
Classification results with the set of multiple parameters from single network is taken as input feature vector.
| Method | Best input | Accuracy | Sensitivity | Specificity |
|---|---|---|---|---|
| vector length | ||||
| CFS | 3 | 0.9259 | 0.9434 | 0.9091 |
| DGUFS | 4 | 0.8734 | 0.8772 | 0.8696 |
| Fisher | 2 | 0.8969 | 0.9259 | 0.8696 |
| FSV | 3 | 0.9341 | 0.9379 | 0.9321 |
| LLCFS | 6 | 0.9009 | 0.9091 | 0.8929 |
| mRMR | 5 | 0.7937 | 0.7692 | 0.8197 |