Table 5.
The variation in discrete outcome measures for different speeds in subject 3.
| Parameter | Speed | Reference (IMU) | Estimate | ||
|---|---|---|---|---|---|
| Left | Right | Left | Right | ||
| Max knee flexion (degrees) | 10 km/h | 34.49 (1.10) | 33.90 (1.20) | 30.57 (0.98) | 30.24 (1.22) |
| 12 km/h | 35.64 (1.20) | 34.71 (1.13) | 36.18 (0.71) | 34.92 (0.98) | |
| 14 km/h | 36.93 (1.24) | 35.11 (1.16) | 38.22 (1.58) | 36.23 (2.76) | |
| peak vGRF (BW) | 10 km/h | 2.85 (0.06) | 2.77 (0.06) | 2.76 (0.07) | 2.67 (0.08) |
| 12 km/h | 3.00 (0.06) | 2.90 (0.07) | 3.06 (0.04) | 2.93 (0.05) | |
| 14 km/h | 3.13 (0.07) | 3.00 (0.07) | 2.96 (0.13) | 2.92 (0.10) | |
| Loading rate (BW/s) | 10 km/h | 52.92 (5.82) | 55.11 (6.10) | 46.05 (5.92) | 56.05 (8.37) |
| 12 km/h | 55.47 (6.12) | 61.34 (6.29) | 47.67 (6.28) | 56.13 (6.03) | |
| 14 km/h | 63.25 (7.31) | 67.17 (9.78) | 59.11 (9.06) | 55.52 (13.91) | |
The mean (and standard deviation) of peak vGRF, loading rate and max knee flexion during stance are shown for both the estimate and its corresponding reference (based on inertial full-body motion capture data), these are calculated over approximately 400 steps (left and right combined). The artificial neural networks were trained using running data of two speeds (different from the evaluation speed), while the shown speed was used for evaluation.