Table 2.

MSE, $E_{\mathrm{p}}$, $R^2$, and AUC values for dataset II from different motion reduction algorithms and no correction. The processing time is calculated for 36 channels (6.6 min of recording with 50-Hz sampling rate) on a 3.4 Hz CPU running windows 7. The bold values are the best values among all algorithms. The proposed method is highlighted in italics.

Dataset II	$\mathrm{MSE}\times {10}^4$	$E_{\mathrm{p}}\times {10}^4$	$R^2$	AUC	Processing time (s)
No correction	$0.75\pm 0.11$	$4.34\pm 0.73$	$0.79\pm 0.02$	$0.88\pm 0.05$	—
Spline	$0.73\pm 0.11$	$4.24\pm 0.72$	$0.79\pm 0.02$	$\mathbf{0.89}\pm \mathbf{0.05}$	2
Wavelet	$0.55\pm 0.08$	$2.97\pm 0.49$	$0.81\pm 0.01$	$0.88\pm 0.05$	529
CBSI	$0.54\pm 0.09$	$3.39\pm 0.56$	$\mathbf{0.84}\pm \mathbf{0.01}$	$\mathrm{0.84}\pm \mathrm{0.01}$	0.6
tPCA	$0.76\pm 0.12$	$4.31\pm 0.72$	$0.74\pm 0.02$	$0.85\pm 0.06$	10
tPCA-SG	$0.51\pm 0.07$	$3.10\pm 0.54$	$0.82\pm 0.01$	$0.86\pm 0.07$	46
SG	$0.45\pm 0.06$	$2.53\pm 0.41$	$0.83\pm 0.01$	$0.88\pm 0.05$	0.1
Spline-SG	0.44 ± 0.06	2.52 ± 0.41	0.83 ± 0.01	0.89 ± 0.05	16
Rloess	$0.72\pm 0.10$	$4.26\pm 0.72$	$0.79\pm 0.02$	$0.87\pm 0.05$	1800
Spline-Rloess	$0.56\pm 0.08$	$3.36\pm 0.59$	$0.82\pm 0.01$	$\mathbf{0.89}\pm \mathbf{0.05}$	1800