A multi-type features fusion neural network for blood pressure prediction based on photoplethysmography. (July 2021)

• Record Type: Journal Article
• Authors: Rong, Meng; Li, Kaiyang
• Is Part Of: Biomedical signal processing and control. Volume 68(2021)
• Publication Date: 2021-07
• Subjects: MTFF multi-type features fusion -- CNN convolutional neural network -- SVM support vector machine -- UCI UC Irvine -- BP blood pressure -- STD standard deviation -- MAE mean absolute error -- SBP systolic blood pressure -- DBP diastolic blood pressure -- ROI region of interest -- ML machine learning -- CVD cardiovascular disease -- PWV pulse wave velocity -- PPG photoplethysmography -- PTT pulse transit time -- DNN deep neural network -- ANN artificial neural network -- ABP arterial blood pressure -- LSTM long short term memory -- RNN recurrent neural network -- ICU intensive care unit -- ME mean bias -- BHS British Hypertension Society -- AAMI Advancement of Medical Instrumentation -- MIMC Medical Information Mart for Intensive Care; Multi-type features fusion -- Blood pressure (BP) -- Photoplethysmography (PPG) -- Deep learning; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.102772
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: This paper proposed a multi-type features fusion (MTFF) neural network for blood pressure (BP) prediction based on photoplethysmography (PPG). The model includes two convolutional neural networks (CNN) and one Bi-directional long short term memory (BLSTM) network. Among them, two CNN networks are used to train the morphological and frequency spectrum features of PPG signal, and the BLSTM network is used to train the temporal features of PPG signal. Compared with the traditional manual calculation features of blood pressure prediction method, our method automatically extracts PPG features through the deep learning model and avoids the error of manually calculating. With the training of multiple features, the deep learning model can obtain more information of PPG signals. Blood pressure prediction based on the fused features further improves accuracy. Our model only needs PPG signal to predict blood pressure. Compared to other works that require many different type of biological signals, a single PPG signal is more convenient to obtain. Abstract: Blood pressure monitoring is very important for the prevention of cardiovascular diseases. In this paper, we proposed a multi-type features fusion (MTFF) neural network model for blood pressure (BP) prediction based on photoplethysmography (PPG). The model includes two convolutional neural networks (CNN) which used to train the morphological and frequency spectrum features of PPG signal, and one Bi-directional long short term memory (BLSTM) network which used to train the temporal features of PPG signal. These multi-features were fused through a specific fusion module after training, so more information of PPG signals were obtained and the hidden relationship between the fused features and blood pressure was established. The standard deviation (STD) and mean absolute error (MAE) of the fusion model are 7.25 mmHg and 5.59 mmHg respectively for systolic blood pressure (SBP), 4.48 mmHg and 3.36 mmHg respectively for diastolic blood pressure (DBP). The results are in full compliance with the Advancement of Medical Instrumentation (AAMI) and the British Hypertension Society (BHS) international standards. We conclude that the MTFF neural network proposed in this paper can accurately predict blood pressure. The significant difference from the traditional methods of BP prediction based on manual calculation of features is that our method automatically extracts PPG features through the deep learning model which can easily handle the complicated and tedious calculation. Compared with other similar BP prediction methods based on deep learning, three different features are trained and fused, which further improves the accuracy of BP prediction.