Continuous Blood Pressure Estimation From Electrocardiogram and Photoplethysmogram During Arrhythmias

Abstract

Objective: Continuous blood pressure (BP) provides valuable information for the disease management of patients with arrhythmias. The traditional intra-arterial method is too invasive for routine healthcare settings, whereas cuff-based devices are inferior in reliability and comfortable for long-term BP monitoring during arrhythmias. The study aimed to investigate an indirect method for continuous and cuff-less BP estimation based on electrocardiogram (ECG) and photoplethysmogram (PPG) signals during arrhythmias and to test its reliability for the determination of BP using invasive BP (IBP) as reference.

Methods: Thirty-five clinically stable patients (15 with ventricular arrhythmias and 20 with supraventricular arrhythmias) who had undergone radiofrequency ablation were enrolled in this study. Their ECG, PPG, and femoral arterial IBP signals were simultaneously recorded with a multi-parameter monitoring system. Fifteen features that have the potential ability in indicating beat-to-beat BP changes during arrhythmias were extracted from the ECG and PPG signals. Four machine learning algorithms, decision tree regression (DTR), support vector machine regression (SVR), adaptive boosting regression (AdaboostR), and random forest regression (RFR), were then implemented to develop the BP models.

Results: The results showed that the mean value ± standard deviation of root mean square error for the estimated systolic BP (SBP), diastolic BP (DBP) with the RFR model against the reference in all patients were 5.87 ± 3.13 and 3.52 ± 1.38 mmHg, respectively, which achieved the best performance among all the models. Furthermore, the mean error ± standard deviation of error between the estimated SBP and DBP with the RFR model against the reference in all patients were -0.04 ± 6.11 and 0.11 ± 3.62 mmHg, respectively, which complied with the Association for the Advancement of Medical Instrumentation and the British Hypertension Society (Grade A) standards.

Conclusion: The results indicated that the utilization of ECG and PPG signals has the potential to enable cuff-less and continuous BP estimation in an indirect way for patients with arrhythmias.

Keywords: arrhythmias; continuous blood pressure; electrocardiogram; machine learning algorithms; photoplethysmogram.

FIGURE 1

(A) Experimental scene of simultaneous acquisition of ECG, PPG, and IBP signals. (B) Typical ECG, PPG, and IBP waveforms recorded in arrhythmias. The type of each beat is labeled in the ECG waveforms. The two adjacent beats in the IBP waveforms with the largest changes in SBP are marked in the red font. ECG, electrocardiogram; IBP, invasive blood pressure; PVC, premature ventricular contraction; PPG, photoplethysmogram; SR, sinus rhythm; SBP, systolic blood pressure.

FIGURE 2

Extracted physiological features from ECG, PPG, and 1st dPPG in a cardiac cycle. First dPPG indicates the first derivative of PPG. (A) PTT values extraction. (B) PPG features extraction. R represent the R-wave peak of ECG; F/F′ and P represent the foot and peak of the PPG waveform, respectively, and letter M (and its corresponding point M′) represents the peak of the first derivative of PPG. Abbreviations and detailed definitions show in Table 1.

FIGURE 3

Dataset partition for each patient (A) and flowchart of blood pressure estimation model construction and evaluation (B). RMSE_DTR, RMSE_SVR, RMSE_AdaboostR, and RMSE_RFR represents the root mean square error between the reference and the estimated blood pressure values by the DTR, SVR, AdaboostR, and RFR, respectively. RMSE, root-mean-square error; DTR, decision tree regression; SVR, support vector machine regression; AdaboostR, adaptive boosting regression; RFR, random forest regression.

FIGURE 4

(A) Performance (mean value ± standard deviation of RMSEs) of different regression algorithms to estimate SBP and DBP in the validation set of all patients. Triple asterisks “*,” “**,” and “ns” indicate statistical significance at p < 0.05, p < 0.01, and p > 0.05, respectively. (B) Estimated beat-to-beat SBP and DBP comparisons in a representative patient, with the proposed method indicated in red and the reference shown in black. Abbreviations show in Figure 3.

FIGURE 5

Correlation and Bland–Altman plots for estimated SBP (A,B) and DBP (C,D) values from the proposed model (by using the random forest regression model) versus the references in a representative patient. In (B) and (D), the black dotted and solid red lines represent the ME ± 1.96 × STD. DBP, diastolic blood pressure; ME, mean error; STD, standard deviation of error; SBP, systolic blood pressure.

FIGURE 6

Group average absolute correlations (mean value ± standard deviation) between three PTTs and the PPG features that most relevant to blood pressure versus (A) SBP and (B) DBP. DBP, diastolic blood pressure; SBP, systolic blood pressure; other abbreviations and detailed definitions show in Table 1.