Beat-to-beat continuous blood pressure estimation with optimal feature set of PPG and ECG signals using deep recurrent neural networks

Abstract

Aim: Continuous blood pressure (BP) monitoring can provide invaluable information for cardiovascular disease (CVD) diagnosis. The purpose of this study is to develop a deep recurrent neural network (RNN) model with an optimal feature set of photoplethysmogram (PPG) and electrocardiogram (ECG) signals for continuous BP estimation.

Methods: This paper presents a novel deep recurrent neural network (RNN), which consists of 2-layered bidirectional Long Short-term Memory (Bi-LSTM) and 6-layered LSTM networks. It is used to estimate BP based on the optimal feature set of PPG and ECG signals. In this work, the optimal feature set is determined using five different feature selection methods.

Results: The proposed method is evaluated based on 660 subjects from the University of California Irvine (UCI) machine learning repository. The RNN model with optimal feature set achieved root mean square error (RMSE) of 3.223 and 1.781 mmHg for systolic BP (SBP) and diastolic BP (DBP), respectively. It also showed mean absolute error (MAE) of 2.514 and 1.383 mmHg for SBP and DBP, respectively. Regarding the British Hypertension Society (BHS) standard, the results attained grade A for the estimation of SBP and DBP.

Conclusion: The experimental results suggest that the proposed deep RNN model with an optimal feature set can improve the performance of BP prediction. Thus, it is possible to further apply our proposed method to develop a wearable device for real-time BP monitoring.

Keywords

Continuous blood pressure / photoplethysmogram / electrocardiogram / feature selection / recurrent neural network

References

[1] Liu Q,Zhang Y.Beats-to-beats estimation of blood pressure during supine cycling exercise using a probabilistic nonparametric method.IEEE Access2021;9:115655-63

[2] World health statistics 2015. Geneva: World Health Organization; 2015. Avaliable from: https://www.who.int/publications/i/item/9789240694439 [Last accessed on 16 Oct 2023]

[3] Chow CK,Rangarajan S.PURE (Prospective Urban Rural Epidemiology) Study investigatorsPrevalence, awareness, treatment, and control of hypertension in rural and urban communities in high-, middle-, and low-income countries.JAMA2013;310:959-68

[4] Thambiraj G,Mangalanathan U,Anand M.Investigation on the effect of Womersley number, ECG and PPG features for cuff less blood pressure estimation using machine learning.Biomed Signal Process Control2020;60:101942

[5] Simjanoska M,Bogdanova AM,Gams M.ECG-derived blood pressure classification using complexity analysis-based machine learning. Proceedings of the 11th International Joint Conference on Biomedical Engineering Systems and Technologies, 2018 Jan 19-21; Funchal, Portugal.

[6] Zhou ZB,Li D.Wearable continuous blood pressure monitoring devices based on pulse wave transit time and pulse arrival time: a review.Materials2023;16:2133 PMCID:PMC10057755

[7] Ding XR,Yang GZ.Continuous blood pressure measurement from invasive to unobtrusive: celebration of 200th birth anniversary of Carl Ludwig.IEEE J Biomed Health Inform2016;20:1455-65

[8] Poon CC.Cuff-less and noninvasive measurements of arterial blood pressure by pulse transit time.Conf Proc IEEE Eng Med Biol Soc2005;2005:5877-80

[9] Ding XR,Liu J,Tsang HK.Continuous cuffless blood pressure estimation using pulse transit time and photoplethysmogram intensity ratio.IEEE Trans Biomed Eng2016;63:964-72

[10] Yan WR,Zhang YT.Cuffless continuous blood pressure estimation from pulse morphology of photoplethysmograms.IEEE Access2019;7:141970-7

[11] Esmaelpoor J,Kadkhodamohammadi A.A multistage deep neural network model for blood pressure estimation using photoplethysmogram signals.Comput Biol Med2020;120:103719

[12] Saeed M,Reisner AT.Multiparameter intelligent monitoring in intensive care II: a public-access intensive care unit database.Crit Care Med2011;39:952-60

[13] Kachuee M,Mohammadzade H.Cuff-less high-accuracy calibration-free blood pressure estimation using pulse transit time. Proceedings of IEEE international symposium on circuits and systems (ISCAS); 2015 May 24-27; Lisbon, Portugal.

[14] Chen H,Morgan J.An effective PSR-based arrhythmia classifier using self-similarity analysis.Biomed Signal Process Control2021;69:102851

[15] Chen H.An automatic R-peak detection method based on hierarchical clustering. Proceedings of IEEE Biomedical Circuits and Systems Conference (BioCAS); 2019 Oct 17-19; Nara, Japan.

[16] Teng XF.Continuous and noninvasive estimation of arterial blood pressure using a photoplethysmographic approach. Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No. 03CH37439); 2003 Sep 17-21; Cancun, Mexico.

[17] Khaire UM.Stability of feature selection algorithm: a review.Comput Inf Sci2022;34:1060-73

[18] Robnik-šikonja M.Theoretical and empirical analysis of ReliefF and RReliefF.Mach Learn2003;53:23-69

[19] Zhang L,Yang F.Noninvasive continuous blood pressure estimation with fewer parameters based on RA-ReliefF feature selection and MPGA-BPN models.Biomed Signal Process Control2023;84:104757

[20] Raju STU,Hossain MI.A novel technique for continuous blood pressure estimation from optimal feature set of ppg signal using deep learning approach. Res Sq Forthcoming 2023

[21] Kavsaoğlu AR,Hariharan M.Non-invasive prediction of hemoglobin level using machine learning techniques with the PPG signal's characteristics features.Appl Soft Comput2015;37:983-91

[22] Vergara JR.A review of feature selection methods based on mutual information.Neural Comput Appl2014;24:175-86

[23] Goldberger J,Roweis S. Neighbourhood components analysis. In: Saul LK, Weiss Y, Bottou L, editors. Advances in neural information processing systems 17. Cambridge: The MIT Press; 2005.

[24] Raghu S.Classification of focal and non-focal EEG signals using neighborhood component analysis and machine learning algorithms.Expert Syst Appl2018;113:18-32

[25] Amodei D,Anubhai R.Deep speech 2: end-to-end speech recognition in English and Mandarin. Proceedings of The 33rd International Conference on Machine Learning; 2016 Jun 19-24; New York, USA.

[26] Li S,Cook C,Gao Y.Independently recurrent neural network (indrnn): Building a longer and deeper rnn. Proceedings of the IEEE conference on computer vision and pattern recognition; 2018 Jun 18-22; Salt Lake City, USA.

[27] He K,Ren S.Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition; 2016 Jun 27-30; Las Vegas, USA.

[28] Karun KM.BA.plot: an R function for bland-altman analysis.Clin Epidemiol Glob Health2021;12:100831

[29] Qiu Y,Yang G.Cuffless blood pressure estimation based on composite neural network and graphics information.Biomed Signal Process Control2021;70:103001

[30] El-hajj C.Deep learning models for cuffless blood pressure monitoring from PPG signals using attention mechanism.Biomed Signal Process Control2021;65:102301

[31] Ganti VG,Nevius BN,Etemadi M.Wearable cuff-less blood pressure estimation at home via pulse transit time.IEEE J Biomed Health Inform2021;25:1926-37 PMCID:PMC8221527

[32] O'Brien E,Littler W.The British hypertension society protocol for the evaluation of automated and semi-automated blood pressure measuring devices with special reference to ambulatory systems.J Hypertens1990;8:607-19