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

Table 21

DBP prediction models for 16-second epoching.

Info				Performance evaluation criteria
L	FN	FP	Model	MAPE	MAD	SE	MSE	RMSE
1	1	5	FT	3.71	2.81	4.17	17.33	4.16	0.94	0.88
			GPR	3.33	2.50	3.72	13.84	3.72	0.94	0.89
			EBT	3.46	2.61	3.88	15.02	3.88	0.94	0.88
2	3	10	FT	3.62	2.76	4.04	16.32	4.04	0.95	0.89
			GPR	3.87	2.96	19.45	377.68	19.43	0.94	0.89
			EBT	3.34	2.48	3.86	14.86	3.85	0.95	0.90
3	4	15	FT	3.62	2.76	4.04	16.32	4.04	0.95	0.89
			GPR	5.93	4.62	76.65	5865.87	76.59	0.94	0.89
			EBT	3.34	2.50	3.79	14.34	3.79	0.95	0.90
4	5	20	FT	3.62	2.76	4.04	16.32	4.04	0.95	0.89
			GPR	7.76	6.09	127.74	16289.21	127.63	0.94	0.89
			EBT	3.35	2.50	3.87	14.93	3.86	0.95	0.90
5	6	25	FT	3.31	2.52	4.03	16.23	4.03	0.95	0.91
			GPR	6.20	4.84	84.13	7066.21	84.06	0.94	0.89
			EBT	3.19	2.36	3.76	14.09	3.75	0.96	0.91
6	8	30	FT	3.46	2.62	4.13	17.06	4.13	0.95	0.90
			GPR	21.04	16.78	499.07	248655.06	498.65	0.95	0.90
			EBT	3.04	2.26	3.59	12.87	3.59	0.96	0.92
7	9	35	FT	3.45	2.62	4.13	17.03	4.13	0.95	0.90
			GPR	20.87	16.64	493.67	243303.01	493.26	0.95	0.90
			EBT	3.08	2.29	3.69	13.56	3.68	0.96	0.92
8	10	40	FT	3.37	2.55	4.07	16.55	4.07	0.95	0.90
			GPR	5.70	4.41	67.87	4599.29	67.82	0.95	0.89
			EBT	3.06	2.28	3.66	13.41	3.66	0.96	0.92
9	11	45	FT	3.37	2.55	4.07	16.55	4.07	0.95	0.90
			GPR	4.36	3.24	4.94	24.34	4.93	0.88	0.78
			EBT	3.10	2.30	3.72	13.80	3.72	0.96	0.92
10	13	50	FT	3.06	2.32	3.74	13.95	3.73	0.96	0.92
			GPR	3.01	2.26	3.59	12.89	3.59	0.96	0.92
			EBT	2.85	2.11	3.51	12.30	3.51	0.97	0.93
11	25	100	FT	2.61	1.99	3.71	13.73	3.71	0.97	0.95
			GPR	2.44	1.83	3.11	9.64	3.10	0.98	0.95
			EBT	2.44	1.81	3.23	10.42	3.23	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.