A multi-label deep residual shrinkage network for high-density surface electromyography decomposition in real-time

Table 6 Decomposition accuracy of MT-DCNN for different window sizes and step sizes

Window size (data points)	Step size (data points)	Precision	Sensitivity	F1-score	Correctly identified MU counts
20	10	0.66 ± 0.16	0.45 ± 0.16	0.51 ± 0.17	0.92 ± 0.16
	20	0.51 ± 0.19	0.29 ± 0.17	0.35 ± 0.19	0.13 ± 0.44
	30	0.36 ± 0.21	0.19 ± 0.15	0.23 ± 0.17	0.02 ± 0.14
	40	0.29 ± 0.20	0.13 ± 0.12	0.16 ± 0.15	0.00 ± 0.00
	50	0.23 ± 0.18	0.10 ± 0.09	0.12 ± 0.12	0.00 ± 0.00
60	10	0.93 ± 0.05	0.86 ± 0.06	0.89 ± 0.05	5.46 ± 5.07
	20	0.88 ± 0.07	0.79 ± 0.09	0.83 ± 0.08	3.85 ± 4.17
	30	0.79 ± 0.12	0.70 ± 0.13	0.74 ± 0.12	2.60 ± 3.44
	40	0.70 ± 0.17	0.60 ± 0.18	0.64 ± 0.18	1.56 ± 2.32
	50	0.63 ± 0.16	0.53 ± 0.18	0.56 ± 0.18	0.79 ± 1.44
100	10	0.95 ± 0.05	0.90 ± 0.05	0.92 ± 0.05	8.96 ± 6.89
	20	0.92 ± 0.06	0.86 ± 0.07	0.89 ± 0.06	6.50 ± 5.94
	30	0.86 ± 0.09	0.80 ± 0.10	0.83 ± 0.09	4.44 ± 4.68
	40	0.80 ± 0.12	0.76 ± 0.13	0.77 ± 0.13	3.63 ± 3.94
	50	0.74 ± 0.13	0.72 ± 0.15	0.72 ± 0.14	2.79 ± 3.22
140	10	0.96 ± 0.04	0.92 ± 0.04	0.94 ± 0.04	11.90 ± 8.44
	20	0.93 ± 0.05	0.90 ± 0.06	0.91 ± 0.05	9.69 ± 7.14
	30	0.88 ± 0.08	0.86 ± 0.09	0.87 ± 0.08	7.73 ± 6.41
	40	0.83 ± 0.11	0.83 ± 0.13	0.83 ± 0.12	6.77 ± 5.40
	50	0.79 ± 0.12	0.79 ± 0.16	0.78 ± 0.14	6.21 ± 5.31

Values are expressed as mean ± standard deviation across 16 participants and 3 contraction intensities

ISSN: 1743-0003