Real-Time Cuffless Continuous Blood Pressure Estimation Using Deep Learning Model

Abstract

Blood pressure monitoring is one avenue to monitor people’s health conditions. Early detection of abnormal blood pressure can help patients to get early treatment and reduce mortality associated with cardiovascular diseases. Therefore, it is very valuable to have a mechanism to perform real-time monitoring for blood pressure changes in patients. In this paper, we propose deep learning regression models using an electrocardiogram (ECG) and photoplethysmogram (PPG) for the real-time estimation of systolic blood pressure (SBP) and diastolic blood pressure (DBP) values. We use a bidirectional layer of long short-term memory (LSTM) as the first layer and add a residual connection inside each of the following layers of the LSTMs. We also perform experiments to compare the performance between the traditional machine learning methods, another existing deep learning model, and the proposed deep learning models using the dataset of Physionet’s multiparameter intelligent monitoring in intensive care II (MIMIC II) as the source of ECG and PPG signals as well as the arterial blood pressure (ABP) signal. The results show that the proposed model outperforms the existing methods and is able to achieve accurate estimation which is promising in order to be applied in clinical practice effectively.

An Estimation Method of Continuous Non-Invasive Arterial Blood Pressure Waveform Using Photoplethysmography: A U-Net Architecture-Based Approach

“…In Carek et al [ 13 ], two subject-specific parameters were also used to calibrate the data, which will vary for different subjects. However, the studies of Li et al [ 12 ] and Esmaelpoor et al [ 47 ] used the publicly available MIMIC-II database. But Li et al [ 12 ] needed two signals (ECG and PPG) and several features to estimate BP.…”

“…However, the studies of Li et al [ 12 ] and Esmaelpoor et al [ 47 ] used the publicly available MIMIC-II database. But Li et al [ 12 ] needed two signals (ECG and PPG) and several features to estimate BP. And on the other hand, our model obtained comparatively better MAE and STD than Esmaelpoor et al [ 47 ].…”

“…Most of the PTT-based methods focus on the combination of two signals, photoplethysmogram (PPG) and electrocardiogram (ECG). In [ 12 ], an ECG and PPG signal feature-based BP estimation process was discussed. The feature set used in the study also included the PTT, that is the time difference between the R wave of ECG and systolic point of PPG.…”

Bed-Based Ballistocardiography: Dataset and Ability to Track Cardiovascular Parameters

“…The idea of continuously monitoring blood pressure using noninvasive means has become a popular topic in recent years [ 7 , 25 , 26 , 27 , 28 ]. Pulse arrival time (PAT), meaning the time delay between the electrical activation of the heart and the arrival of a commensurate pulse wave at a distal point, has shown promise for tracking changes in blood pressure (see [ 25 ] or [ 29 ] for an overview of the relationship between PAT and blood pressure).…”

Beat-to-Beat Continuous Blood Pressure Estimation Using Bidirectional Long Short-Term Memory Network

“…Wu et al proposed a deep neural network (DNN) based on combined information using ECG and PPG [ 14 ]. Moreover, Li et al suggested a real-time BP estimation model with a long short-term memory (LSTM) network using the features of ECG and PPG [ 15 ]. In our previous study, we investigated an end-to-end BP estimation algorithm using a CNN with an attention mechanism [ 16 ].…”