Table 3 Performance analysis of the classification with SVM based on the original WPD features, PCA-based features, and GND-ICA based features in terms of above introduced four measures in percentage
From: Fast multi-scale feature fusion for ECG heartbeat classification
Features | N | S | V | F | MAC | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
AC | Sens | PPV | AC | Sens | PPV | AC | Sens | PPV | AC | Sens | PPV | ||
WPD | 96.68 | 99.00 | 96.54 | 98.30 | 81.80 | 92.14 | 98.06 | 92.70 | 96.49 | 99.40 | 74.56 | 93.15 | 96.21 |
PCA | 96.57 | 98.89 | 96.52 | 98.00 | 76.47 | 92.60 | 98.11 | 94.30 | 95.25 | 99.18 | 67.83 | 87.18 | 95.93 |
GND-ICA | 98.00 | 99.10 | 98.17 | 98.79 | 88.06 | 93.65 | 98.87 | 96.90 | 96.87 | 99.43 | 77.81 | 91.50 | 97.53 |