Research Article
Atrial Fibrillation Beat Identification Using the Combination of Modified Frequency Slice Wavelet Transform and Convolutional Neural Networks
Table 4
The experimental results.
| Fold | Test data | Training data | Acc (%) | Se (%) | Sp (%) | AUC | Acc (%) | Se (%) | Sp (%) | AUC |
| 1 | 86.63 | 77.95 | 95.97 | 0.95 | 97.59 | 96.68 | 98.54 | 0.99 | 2 | 86.82 | 84.28 | 88.62 | 0.93 | 98.21 | 97.81 | 98.62 | 0.99 | 3 | 83.55 | 78.90 | 87.38 | 0.91 | 97.80 | 96.77 | 98.88 | 0.99 | 4 | 65.93 | 58.59 | 72.08 | 0.70 | 98.26 | 98.51 | 98.00 | 0.98 | 5 | 82.41 | 75.06 | 87.99 | 0.90 | 98.40 | 98.56 | 98.24 | 0.99 | Mean | 81.07 | 74.96 | 86.41 | 0.88 | 98.05 | 97.67 | 98.46 | 0.99 | SD | 8.68 | 9.74 | 8.73 | 0.10 | 0.34 | 0.91 | 0.34 | 0 | Mean# | 84.85 | 79.05 | 89.99 | 0.92 | 98.00 | 97.46 | 98.57 | 0.99 | SD# | 1.92 | 3.34 | 3.48 | 0.02 | 0.32 | 0.78 | 0.23 | 0 |
|
|
#The results only from the average of the folds 1, 2, 3, and 5.
|