Table 2.
The difference between linear regression and non-linear neural network for predicting the power loss.
| Linear regression of sEMG-based parameters | Actual versus predicted | Non-linear neural network | Linear versus non-linear | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Slope | Intercept | ||||||||||
| Muscle groups | R2 | Regression equation | SNR | r | r | SNR | F | P | F | P | |
| PFs | MG | 11.02 | Power% = 0.176 × IMNF% + 0.064 × MNF% + 32.473 | 8.590 | 0.218 | 0.353 | 9.040 | 38.89 | 0.000 | 0.08 | 0.780 |
| LG | 5.52 | Power% = 0.134 × IMNF% + 0.068 × MDF% + 33.891 | 7.199 | 0.221 | 0.231 | 8.720 | 15.30 | 0.000 | 0.01 | 0.911 | |
| KEs | VM | 8.24 | Power% = 0.340 × IMNF% − 0.151 × MAV% + 29.999 | 7.668 | 0.353 | 0.361 | 8.244 | 30.54 | 0.000 | 0.05 | 0.816 |
| RF | 21.07 | Power% = 0.679 × IMDF% + 0.073 × MDF% − 0.212 × MAV% + 15.220 | 7.669 | 0.459 | 0.491 | 8.916 | 19.18 | 0.000 | 0.00 | 0.957 | |
| VL | 9.67 | Power% = 0.421 × IMNF% − 0.050 × RMS% + 20.875 | 8.527 | 0.313 | 0.383 | 10.103 | 21.60 | 0.000 | 0.00 | 0.959 | |
| HEs | GM | 13.76 | Power% = 0.626 × IMDF% + 9.881 | 5.555 | 0.355 | 0.437 | 6.692 | 23.20 | 0.000 | 0.00 | 0.996 |
| BF | 23.72 | Power% = 0.290 × IMDF% + 0.414 × MDF% + 5.833 | 5.761 | 0.475 | 0.482 | 11.652 | 23.21 | 0.000 | 0.59 | 0.441 | |
PFs plantar flexors, KEs knee extensors, HEs hip extensors. MG medial gastrocnemius, LG lateral gastrocnemius, VM vastus medialis, RF rectus femoris, VL vastus lateralis, GM gluteus maximus, BF biceps femoris. r pearson's correlation coefficients, R2 coefficient of determination, SNR signal-to-noise ratio. RMS root mean square, MAV mean absolute value, MNF mean frequency, MDF median frequency, IMNF instantaneous mean frequency, IMDF instantaneous median frequency. Changes in percentage of power output (power%) was the dependent variable and changes in percentage of sEMG-based parameters such as RMS%, MAV%, MNF%, MDF%, IMNF%, and IMDF% were the predictor variables in these stepwise multiple regressions.