Table 12 Accuracy of exercise classification on the self-collected dataset using CNN architectures.
From: Spike train analysis in rehabilitation movement classification using deep learning approach
CNN Architecture | Accuracy | Accuracy (mean ± std) | 95% CI | Macro-F1 (mean ± std) | Training time (min/epoch) |
|---|---|---|---|---|---|
CNN163 | 0.9556 | 0.9556 ± 0.005 | [0.947–0.964] | 0.952 ± 0.007 | 2.3 |
CNN264 | 0.8667 | 0.8667 ± 0.008 | [0.854–0.879] | 0.861 ± 0.009 | 3.1 |
CNN365 | 0.9333 | 0.9333 ± 0.006 | [0.922–0.945] | 0.928 ± 0.008 | 2.0 |
CNN465 | 0.9778 | 0.9778 ± 0.004 | [0.970–0.985] | 0.974 ± 0.005 | 5.4 |
Proposed CNN | 1.0000 | 1.0000 ± 0.000 | [1.000–1.000] | 1.000 ± 0.000 | 3.9 |
