Research Article
[Retracted] Automatic Detection of High-Frequency Oscillations Based on an End-to-End Bi-Branch Neural Network and Clinical Cross-Validation
Table 4
The results for the bi-branch feature fusion model using intrasubject validation.
| Group | Confusion matrix | Evaluation metrics (%) | TN | FP | FN | TP | SEN | SPE | PRE | ACC | FDR | F1 | SEN_SPE |
| 1 | 698 | 57 | 34 | 657 | 95.08 | 92.45 | 92.02 | 93.71 | 7.98 | 93.52 | 93.75 | 2 | 692 | 49 | 47 | 631 | 93.07 | 93.39 | 92.79 | 93.23 | 7.21 | 92.93 | 93.23 | 3 | 675 | 69 | 21 | 657 | 96.90 | 90.73 | 90.50 | 93.67 | 9.50 | 93.59 | 93.71 | 4 | 1232 | 130 | 74 | 1126 | 93.83 | 90.46 | 89.65 | 92.04 | 10.35 | 91.69 | 92.11 | 5 | 546 | 20 | 27 | 440 | 94.22 | 96.47 | 95.65 | 95.45 | 4.35 | 94.93 | 95.33 | Average | 94.62 | 92.70 | 92.12 | 93.62 | 7.88 | 93.33 | 93.63 |
|
|