Attention Mechanism-Based Convolutional Long Short-Term Memory Neural Networks to Electrocardiogram-Based Blood Pressure Estimation

Abstract

Monitoring people’s blood pressure can effectively prevent blood pressure-related diseases. Therefore, providing a convenient and comfortable approach can effectively help patients in monitoring blood pressure. In this study, an attention mechanism-based convolutional long short-term memory (LSTM) neural network is proposed to easily estimate blood pressure. To easily and comfortably estimate blood pressure, electrocardiogram (ECG) and photoplethysmography (PPG) signals are acquired. To precisely represent the characteristics of ECG and PPG signals, the signals in the time and frequency domain are selected as the inputs of the proposed NN structure. To automatically extract the features, the convolutional neural networks (CNNs) are adopted as the first part of neural networks. To identify the meaningful features, the attention mechanism is used in the second part of neural networks. To model the characteristic of time series, the long short-term memory (LSTM) is adopted in the third part of neural networks. To integrate the information of previous neural networks, the fully connected networks are used to estimate blood pressure. The experimental results show that the proposed approach outperforms CNN and CNN-LSTM and complies with the Association for the Advancement of Medical Instrumentation standard.

Advances in Cuffless Continuous Blood Pressure Monitoring Technology Based on PPG Signals

“…The prediction results of the model on the two datasets both achieved the grade A of the BHS standard and satisfied the AAMI standard. Chuang et al [ 78 ] screened 11,000 PPG and ECG segments from 45 subjects in the MIMIC dataset and introduced an attention mechanism to the CNN-LSTM model to identify meaningful features. The results showed that combining the time and frequency domain signals of PPG and ECG could fully obtain the intrinsic characteristics of the signals, which resulted in an error (MAE ± STD) of 2.94 ± 4.65 mmHg for SBP and 2.02 ± 3.81 mmHg for DBP.…”

Artificial Intelligence Model for an Electrocardiography-based Blood Pressure Estimation System

“…(3) Some systems developed for non-invasive BP monitoring include photoplethysmography (PPG) optical sensors based on electron waves, a BP estimation device based on the volume compensation principle, a blood modulation magnetic signal mechanism, and a portable cuff BP sensing system. (4)(5)(6)(7)(8)(9)(10)(11) However, these devices are all in the form of stand-alone devices and are only used for BP measurement and do not include other vital signs. In addition, when some of them are used in real life, especially by specific user groups, the ideal laboratory effect cannot be achieved.…”

Development of a Convolutional Neural Network Model to Predict Coronary Artery Disease Based on Single-Lead and Twelve-Lead ECG Signals

“…CAD is commonly identified and diagnosed based on different tests, such as ECG, treadmill ECG, echocardiogram (ECHO), and angiography. The intelligent systems use different neural architectures, such as the convolutional neural network CNN [11], recurrent neural network (RNN) [12], CNN with RNN [13,14], deep belief network (DBN) [3,15], and the fully-connected neural network (FC) [11] to predict electrocardiogram (ECG)-related issues, such as arrhythmia [16][17][18][19][20], atrial fibrillation (AF) [21][22][23], myocardial infarction (MI) [24,25], ST elevation [21], CAD [10,[26][27][28], etc. Tan, J.H.…”