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

Table 25

Literature comparison.

Nu	Ref	Year	Model methods				Diastolic blood pressure performances						Systolic blood pressure performances [b]
			Signal	Features	Method	MAE	MAPE	MAD	MSE	RMSE			MAE	MAPE	MAD	MSE	RMSE	=
1	[33]	2019	PPT-PIR	SSR-CHC	MARS		3.630								7.830
2	[29]	2020	Oscillometric waveforms	Graphical features	WkNN	11.032			200.531	14.161	0.423	0.179	3.520			41.998	6.480	0.948	0.899
3	[31]	2020	Auscultatory and oscillometric waveforms	Time domain	GMM-HMM	2.900							-0.9
4	[34]	2020	PPG-ECG	Chaotic, time, and frequency domain	RNN				1.730	1.240	0.854	0.730				1.210	0.780	0.849	0.720
5	[30]	2020	Oscillometric waveforms	Chaotic, time, and frequency domain	GPR	4.271	0.288		28.843	5.371	0.891	0.794	3.636	0.114		23.845	4.883	0.962	0.925
6	[27]	2020	PPG-ECG	Time domain	RF	5.48				6.000	0.840	0.706	9.000				13.830	0.850	0.723
7	[32]	2020	Speech	Vowels	CNN-R					0.350							0.236
8	[28]	2020	PPG	PPG segment series	CNN-LSTM	3.97					0.950	0.903		0.670				0.950	0.903
9	[26]	2021	Peripheral signals	Hibrit	MLR					3.000	0.970	0.941					3.000	0.970	0.941
10	[20]	2021	PPG	Multitype feature	MTFF-ANN	3.36							5.590
11			Proposed model ECG 2-second	Time domain	EBT		3.310	2.430	16.520	4.060	0.970	0.930		2.580	0.370	25.480	5.050	0.970	0.930
12			Proposed model-ECG 14-second	Time domain	GPR/EBT		3.280	1.870	13.250	3.340	0.980	0.950		2.000	2.680	19.190	4.380	0.980	0.950
13			Proposed model-ECG 16-second	Time domain	GPR		2.440	1.830	9.640	3.100	0.980	0.950		1.920	2.560	16.660	4.080	0.980	0.960

CHC: current heart cycle; CNN-R: convolutional neural networks-regression; EBT: ensemble bagged tree; ECG: electrocardiography; GMM-HMM: Gaussian mixture models and hidden Markov; GPR: Gaussian process regression; LSTM: long-short-term memory; MAD: mean absolute difference; MAE: mean absolute error; MAPE: mean absolute percentage error; MAPE: mean absolute percentage error; MLR: multiple linear regression; MSE: mean square error; MTFF-ANN: multitype feature fusion artificial neural network (2 CNN+1 LSTM); PIR: photoplethysmogram intensity ratio; PPG: photoplethysmography; PPT: pulse transit time; RF: random forest; RMSE: root mean square error; RNN: recurrent neural networks; SE: standard error; SSR: state space reconstruction; MARS: multiadaptive regression spline; WkNN: weighted -near neighbor.