Table 5.
Estimation of heart rate for running by CNNs.
| Speed | Signal | MAE | SDAE | RMSE | CC |
|---|---|---|---|---|---|
| 6.4 km/h | VGG-16 (No Aug) | 7.21 | 8.64 | 11.25 | 0.846 ** |
| VGG-19 (Aug) | 5.24 | 6.72 | 8.52 | 0.917 ** | |
| Ensemble Network | 5.43 | 7.40 | 9.17 | 0.899 ** | |
| 8.5 km/h | VGG-16 (No Aug) | 8.01 | 8.89 | 11.96 | 0.853 ** |
| VGG-19 (Aug) | 5.21 | 7.14 | 8.84 | 0.924 ** | |
| Ensemble Network | 5.94 | 7.64 | 9.67 | 0.908 ** | |
| 10.3 km/h | VGG-16 (No Aug) | 8.62 | 9.81 | 13.05 | 0.824 ** |
| VGG-19 (Aug) | 5.49 | 7.97 | 9.68 | 0.908 ** | |
| Ensemble Network | 6.38 | 8.61 | 10.72 | 0.886 ** |
MAE = mean absolute error; SDAE = standard deviation of absolute error; RMSE = root mean square error; CC = Pearson’s correlation coefficient. Two asterisks represent significant correlation level at p < 0.01. The lowest error and highest correlation values are bolded.