[Retracted] Featureless Blood Pressure Estimation Based on Photoplethysmography Signal Using CNN and BiLSTM for IoT Devices

Table 2

Performance of different approaches.

Model	Input	SBP (mmHg)			DBP (mmHg)
		RMSE	MAE	STD	RMSE	MAE	STD
SVM [12]	5 features from ECG & PPG	—	12.38	16.17	—	6.34	8.45
SVR [31]	35 features from PPG	10.9	8.54	—	5.8	4.34	—
NN [31]	35 features from PPG	11.6	13.4	—	5.9	6.9	—
RF [18]	>15 features from ECG & PPG	13.83	9.54	—	6.80	5.48	—
	Raw PPG signal	13.49	8.92	8.23	8.78	6.14	5.22
Proposed	Raw PPG signal	11.50	7.85	8.41	6.53	4.42	4.80