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

Table 18

DBP prediction models for 10-second epoching.

Info				Performance evaluation criteria
L	FN	FP	Model	MAPE	MAD	SE	MSE	RMSE
1	1	5	FT	6.56	4.86	6.93	48.00	6.93	0.71	0.50
			GPR	5.99	4.42	6.05	36.51	6.04	0.75	0.57
			EBT	6.14	4.52	6.35	40.26	6.35	0.74	0.54
2	3	10	FT	6.04	4.47	6.75	45.55	6.75	0.77	0.59
			GPR	5.85	4.32	5.98	35.71	5.98	0.76	0.58
			EBT	5.51	4.03	5.93	35.13	5.93	0.81	0.65
3	4	15	FT	6.04	4.47	6.75	45.55	6.75	0.77	0.59
			GPR	5.86	4.32	5.98	35.76	5.98	0.76	0.58
			EBT	5.48	4.02	5.92	35.04	5.92	0.81	0.65
4	5	20	FT	6.03	4.46	6.80	46.14	6.79	0.78	0.60
			GPR	5.66	4.15	5.94	35.19	5.93	0.79	0.62
			EBT	5.36	3.92	5.86	34.32	5.86	0.82	0.67
5	6	25	FT	5.34	3.96	6.08	36.90	6.07	0.83	0.69
			GPR	5.68	4.15	5.95	35.36	5.95	0.79	0.62
			EBT	4.89	3.58	5.42	29.31	5.41	0.86	0.74
6	8	30	FT	5.62	4.11	6.51	42.36	6.51	0.82	0.67
			GPR	5.20	3.81	5.63	31.65	5.63	0.83	0.69
			EBT	4.73	3.44	5.36	28.74	5.36	0.87	0.75
7	9	35	FT	5.59	4.08	6.48	41.89	6.47	0.82	0.68
			GPR	5.20	3.80	5.64	31.73	5.63	0.83	0.69
			EBT	4.75	3.46	5.40	29.13	5.40	0.86	0.75
8	10	40	FT	5.65	4.13	6.50	42.24	6.50	0.82	0.67
			GPR	5.21	3.81	5.61	31.48	5.61	0.83	0.68
			EBT	4.74	3.45	5.38	28.95	5.38	0.87	0.75
9	11	45	FT	5.60	4.09	6.48	41.90	6.47	0.82	0.68
			GPR	5.18	3.80	5.60	31.32	5.60	0.83	0.69
			EBT	4.75	3.45	5.39	29.03	5.39	0.87	0.75
10	13	50	FT	3.66	2.73	4.58	21.00	4.58	0.95	0.90
			GPR	3.27	2.40	3.84	14.71	3.84	0.96	0.92
			EBT	3.23	2.36	3.87	14.94	3.86	0.96	0.92
11	25	100	FT	3.05	2.30	4.16	17.32	4.16	0.96	0.93
			GPR	2.85	2.08	3.53	12.46	3.53	0.97	0.94
			EBT	2.69	1.96	3.42	11.67	3.42	0.97	0.95

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