Table 1. Parameters of four group standards in stage N2 and N3 sleep data.

Stage	Standard	Number	T-group	T-overlap	$\overline{\text{F1 score}}$	F1 score range	$\overline{\text{F1 score-leave}}$
N2	EGS	5	0.30	0.45	0.84 ± 0.02	0.82 ~ 0.88	0.79 ± 0.03
	nEGS-all	168	0.35	0.30	0.72 ± 0.10	0.24 ~ 0.88	0.70 ± 0.10
	nEGS-1	168	0.35	0.30	0.68 ± 0.11	0.24 ~ 0.90	0.66 ± 0.11
	nEGS-05	168	0.05	0.30	0.27 ± 0.13	0.003 ~ 0.64	0.17 ± 0.08
N3	EGS	5	0.30	0.40	0.74 ± 0.05	0.66 ~ 0.79	0.63 ± 0.06
	nEGS-all	168	0.30	0.30	0.59 ± 0.12	0.22 ~ 0.84	0.55 ± 0.12
	nEGS-1	168	0.35	0.25	0.53 ± 0.14	0.15 ~ 0.79	0.47 ± 0.13
	nEGS-05	168	0.05	0.35	0.26 ± 0.13	0.027 ~ 0.56	0.18 ± 0.09

Data are presented as mean ± standard deviation unless otherwise indicated. The EGS was the expert group standard. The nEGS-all was the non-expert group standard with all spindles when both the weighted value of 1 and 0.5 were considered. The nEGS-1 was the non-expert group standard with definite spindles when the weighted value of 1 was only considered. The nEGS-05 was the non-expert group standard with indefinite spindles when the weighted value of 0.5 was only considered. The Number was the number of experts/non-experts in each standard. The T-group was the best group threshold. The T-overlap was the best overlap threshold. The $\overline{\text{F1 score}}$ was the mean F1 score of individuals in a group related to its own group at the best T-group and T-overlap. The F1 score range was the range of F1 scores of 168 non-experts or five experts. The $\overline{\text{F1 score-leave}}$ was the mean F1 score of all experts or non-experts in the leave-one-out analysis. All standards were established by thresholding the group consensus at the T-group.