Machine Learning and Electrocardiography Signal-Based Minimum Calculation Time Detection for Blood Pressure Detection

Table 9

SBP prediction models for 12-second epoching.

Info				Performance evaluation criteria
L	FN	FP	Model	MAPE	MAD	SE	MSE	RMSE
1	1	5	FT	5.45	7.18	10.02	100.18	10.01	0.71	0.51
			GPR	4.95	6.46	8.61	73.99	8.60	0.77	0.60
			EBT	5.10	6.70	9.11	82.95	9.11	0.74	0.55
2	3	10	FT	4.67	6.18	9.16	83.80	9.15	0.81	0.65
			GPR	4.76	6.25	8.37	69.95	8.36	0.79	0.62
			EBT	4.28	5.60	7.95	63.19	7.95	0.84	0.71
3	4	15	FT	4.67	6.18	9.16	83.80	9.15	0.81	0.65
			GPR	4.76	6.25	8.38	70.11	8.37	0.79	0.62
			EBT	4.21	5.50	7.92	62.60	7.91	0.84	0.71
4	5	20	FT	4.64	6.12	9.01	81.00	9.00	0.81	0.66
			GPR	4.53	5.94	8.21	67.28	8.20	0.81	0.65
			EBT	4.13	5.42	7.83	61.28	7.83	0.84	0.71
5	6	25	FT	4.21	5.57	8.32	69.10	8.31	0.87	0.75
			GPR	3.80	5.01	7.20	51.76	7.19	0.87	0.77
			EBT	3.73	4.90	7.07	49.98	7.07	0.89	0.79
6	8	30	FT	3.79	5.01	7.83	61.25	7.83	0.89	0.79
			GPR	3.58	4.73	6.95	48.30	6.95	0.89	0.79
			EBT	3.35	4.40	6.58	43.26	6.58	0.91	0.83
7	9	35	FT	3.79	5.01	7.83	61.25	7.83	0.89	0.79
			GPR	3.56	4.70	6.89	47.38	6.88	0.89	0.79
			EBT	3.35	4.39	6.61	43.65	6.61	0.91	0.83
8	10	40	FT	3.80	5.02	7.84	61.44	7.84	0.89	0.78
			GPR	3.56	4.70	6.88	47.29	6.88	0.89	0.79
			EBT	3.34	4.39	6.58	43.26	6.58	0.91	0.83
9	11	45	FT	3.80	5.02	7.85	61.48	7.84	0.89	0.78
			GPR	3.56	4.71	6.87	47.07	6.86	0.89	0.79
			EBT	3.31	4.35	6.51	42.26	6.50	0.91	0.83
10	13	50	FT	2.94	3.91	6.09	37.06	6.09	0.95	0.90
			GPR	2.73	3.65	5.31	28.18	5.31	0.95	0.91
			EBT	2.58	3.42	5.16	26.57	5.15	0.96	0.92
11	25	100	FT	2.38	3.19	5.53	30.51	5.52	0.97	0.93
			GPR	2.04	2.73	4.39	19.25	4.39	0.98	0.95
			EBT	2.05	2.73	4.46	19.90	4.46	0.98	0.95

L: level; FN: number of feature; FP: percentage of feature; FT: fine tree; GPR: Gaussian process regression; EBT: ensemble bagged tree.