Table 5 Performance of the 9-class classification model.
From: A hybrid model combining 1D-CNN and BERT for intelligent ECG arrhythmia classification
Class | Accuracy/% | Specificity/% | Precision/% | Recall/% | F1 score/% |
|---|---|---|---|---|---|
C1 | 99.33 | 98.46 | 99.33 | 99.71 | 99.52 |
C2 | 99.51 | 99.96 | 99.50 | 99.75 | 99.63 |
C3 | 99.78 | 99.98 | 99.78 | 99.40 | 99.59 |
C4 | 94.93 | 99.89 | 94.93 | 90.75 | 92.79 |
C5 | 97.27 | 99.81 | 97.27 | 98.19 | 97.73 |
C6 | 92.59 | 99.98 | 92.59 | 84.75 | 88.50 |
C7 | 95.12 | 99.97 | 95.12 | 73.58 | 82.98 |
C8 | 98.89 | 98.99 | 98.89 | 94.68 | 96.74 |
C9 | 99.61 | 99.97 | 99.62 | 99.39 | 99.50 |
