Table 14 Comparison of DL prediction and firing rate calculation correctness metrics on the UI-PRMD dataset.
From: Spike train analysis in rehabilitation movement classification using deep learning approach
Exercise | DL Prediction | Firing rate calculation | |||||
|---|---|---|---|---|---|---|---|
Correctness prediction | Precision | Recall | F1-score | Mean percentage error | Mean absolute deviation | Movement correctness | |
Deep squat | 0.8396 | 0.8396 | 0.8676 | 0.8521 | 0.2235 | 0.4825 | 0.7765 |
Hurdle step | 0.7591 | 0.7536 | 0.7803 | 0.7666 | 0.2052 | 0.7104 | 0.7948 |
Inline lunges | 0.7236 | 0.7183 | 0.7552 | 0.7363 | 0.3269 | 0.6951 | 0.6731 |
Sit to stand | 0.8105 | 0.8102 | 0.8238 | 0.8169 | 0.1685 | 0.5348 | 0.8315 |
Shoulder extension | 0.7957 | 0.7963 | 0.8166 | 0.8048 | 0.2161 | 0.7142 | 0.7839 |
Leg raise | 0.7445 | 0.7459 | 0.7718 | 0.7586 | 0.2656 | 0.6839 | 0.7344 |