A Smart Detection Method of Sleep Quality Using EEG Signal and Long Short-Term Memory Model

Table 4

Sleep staging results of the test dataset after adding noise.

Test sample	Noise	W	S1	S2	S3	REM	Accuracy
SC03	(0,0.1)	178	51	364	114	85	0.8752
	(0,0.2)	175	49	363	112	84	0.8596
	(0,0.3)	173	47	360	110	82	0.8421
	(0,0.4)	171	44	359	108	80	0.8220
	(0,0.5)	168	42	357	107	76	0.8015
SC04	(0,0.1)	40	73	360	140	171	0.8520
	(0,0.2)	39	71	359	138	170	0.8395
	(0,0.3)	37	69	358	136	169	0.8229
	(0,0.4)	36	67	357	135	168	0.8116
	(0,0.5)	34	67	355	133	168	0.8001
ST03	(0,0.1)	40	56	298	248	138	0.8518
	(0,0.2)	39	55	298	247	136	0.8415
	(0,0.3)	39	54	297	246	136	0.8373
	(0,0.4)	38	54	294	245	135	0.8295
	(0,0.5)	37	52	293	244	134	0.8169
ST04	(0,0.1)	17	54	420	121	197	0.8706
	(0,0.2)	17	53	419	119	196	0.8632
	(0,0.3)	16	52	418	119	196	0.8502
	(0,0.4)	15	51	418	118	196	0.8361
	(0,0.5)	14	50	418	117	195	0.8212

0 represents mean and 0.1 represents variance.