Abstract
The pulse arrival time (PAT), the difference between the R-peak time of electrocardiogram (ECG) signal and the systolic peak of photoplethysmography (PPG) signal, is an indicator that enables noninvasive and continuous blood pressure estimation. However, it is difficult to accurately measure PAT from ECG and PPG signals because they have inconsistent shapes owing to patient-specific physical characteristics, pathological conditions, and movements. Accordingly, complex preprocessing is required to estimate blood pressure based on PAT. In this paper, as an alternative solution, we propose a noninvasive continuous algorithm using the difference between ECG and PPG as a new feature that can include PAT information. The proposed algorithm is a deep CNN–LSTM-based multitasking machine learning model that outputs simultaneous prediction results of systolic (SBP) and diastolic blood pressures (DBP). We used a total of 48 patients on the PhysioNet website by splitting them into 38 patients for training and 10 patients for testing. The prediction accuracies of SBP and DBP were 0.0 ± 1.6 mmHg and 0.2 ± 1.3 mmHg, respectively. Even though the proposed model was assessed with only 10 patients, this result was satisfied with three guidelines, which are the BHS, AAMI, and IEEE standards for blood pressure measurement devices.
Highlights
There are two types of methods used to measure blood pressure: invasive and noninvasive
By conducting the Durbin–Watson test to verify the autocorrelation between the observed values and the predicted values of the proposed model, it was confirmed that independence of the systolic blood pressure (SBP) and diastolic blood pressure (DBP) errors was satisfied (d-statics = 1.97 for SBP and 1.99 for DBP)
We developed an noninvasive blood pressure (NIBP) algorithm using a combined deep CNN–LSTM network-based multitasking learning architecture
Summary
There are two types of methods used to measure blood pressure: invasive and noninvasive. Many research groups have proposed a blood pressure measurement algorithm based on electrocardiography (ECG) and photoplethysmography (PPG) for noninvasive and continuous blood pressure measurements[4,5]. Chen et al proposed a blood pressure estimation method using a genetic algorithm-mean influence value-support vector regression (GA-MIV-SVR) They extracted various features, including features related to PAT from ECG and PPG signals, and selected features to predict systolic blood pressure (SBP) and diastolic blood pressure (DBP) using mean influence value rankings. Kachuee et al extracted the heart rate, PPG features, and PAT features from ECG and PPG through feature engineering and used them to continuously estimate blood pressure, considering changes in PAT according to individual p hysiology[5] They showed that DBP can be accurately estimated using a support vector machine method. The proposed algorithm is a combined deep CNN–LSTM network-based multitasking learning architecture model that can predict SBP and DBP simultaneously by considering the morphological features of the ECG and PPG signals, along with temporal features
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
